检索结果-内蒙古大学图书馆

5th International Seminar on Artificial Intelligence, Networking and Information technology, AINIT 2024

作者： Huang, Yiting Zhuang, Xinran Tian, Yifei School of Computer Science and Engineering Nanjing University of Science and Technology Jiangsu Nanjing210094 China

ISBN: (纸本)9798350385557

Point cloud is a kind of 3D data type with location information. Compared with 2D images, it can not only retain the position information of the object, but also retain the depth information, color information and so on. At present, the method based on voxelization is mainly used to process the point cloud, but the voxel partition process ignores the connectivity between points and the uneven partition. With a view to preserve the connectivity between points, scholars try to use point-based methods to process point cloud data, build the point cloud into a graph structure and input it into a graph neural network. At present, the method still has the situation of uneven division, and how to retain the points of small targets is an important problem to be solved. Therefore, we propose to use the farthest point sampling method to obtain more discrete and uniform points, and input the constructed graph into the network with attention mechanism, and use the self-attention and SE(Squeeze-and-Excitation) channel attention respectively to extract the features of interest for identification and positioning. Compared with the baseline model, our model improves the accuracy by 2.07% and 1.54% respectively in the pedestrian and cyclist object evaluation in the moderate difficulty scene. © 2024 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Automatic Transportation Mode Classification Using a Deep Reinforcement Learning Approach With Smartphone Sensors

引用

IEEE Access 2024年 12卷 514-533页

作者： Taherinavid, Siavash Moravvej, Seyed Vahid Chen, Yen-Lin Yang, Jing Ku, Chin Soon Yee, Por Lip Iran University of Science and Technology School of Civil Engineering Tehran13114-16846 Iran Isfahan University of Technology Department of Electrical and Computer Engineering Isfahan84156-83111 Iran National Taipei University of Technology Department of Computer Science and Information Engineering Taipei106344 Taiwan Universiti Malaya Faculty of Computer Science and Information Technology Department of Computer System and Technology Kuala Lumpur50603 Malaysia Universiti Tunku Abdul Rahman Department of Computer Science Kampar31900 Malaysia

The increasing dependence on smartphones with advanced sensors has highlighted the imperative of precise transportation mode classification, pivotal for domains like health monitoring and urban planning. This research is motivated by the pressing demand to enhance transportation mode classification, leveraging the potential of smartphone sensors, notably the accelerometer, magnetometer, and gyroscope. In response to this challenge, we present a novel automated classification model rooted in deep reinforcement learning. Our model stands out for its innovative approach of harnessing enhanced features through artificial neural networks (ANNs) and visualizing the classification task as a structured series of decision-making events. Our model adopts an improved differential evolution (DE) algorithm for initializing weights, coupled with a specialized agent-environment relationship. Every correct classification earns the agent a reward, with additional emphasis on the accurate categorization of less frequent modes through a distinct reward strategy. The Upper Confidence Bound (UCB) technique is used for action selection, promoting deep-seated knowledge, and minimizing reliance on chance. A notable innovation in our work is the introduction of a cluster-centric mutation operation within the DE algorithm. This operation strategically identifies optimal clusters in the current DE population and forges potential solutions using a pioneering update mechanism. When assessed on the extensive HTC dataset, which includes 8311 hours of data gathered from 224 participants over two years. Noteworthy results spotlight an accuracy of 0.88±0.03 and an F-measure of 0.87±0.02, underscoring the efficacy of our approach for large-scale transportation mode classification tasks. This work introduces an innovative strategy in the realm of transportation mode classification, emphasizing both precision and reliability, addressing the pressing need for enhanced classification mechanisms in an eve

关键词： Smartphones

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral Image Classification Based on A Multi-Scale Weighted Kernel Network

引用

Chinese Journal of Electronics 2022年第5期31卷 832-843页

作者： SUN Le XU Bin LU Zhenyu School of Computer Science Nanjing University of Information Science and Technology Engineering Research Center of Digital Forensics Ministry of Education Nanjing University of Information Science and Technology School of Artificial Intelligence Nanjing University of Information Science and Technology

Recently, many deep learning models have shown excellent performance in hyperspectral image(HSI) classification. Among them, networks with multiple convolution kernels of different sizes have been proved to achieve richer receptive fields and extract more representative features than those with a single convolution kernel. However, in most networks, different-sized convolution kernels are usually used directly on multibranch structures, and the image features extracted from them are fused directly and simply. In this paper, to fully and adaptively explore the multiscale information in both spectral and spatial domains of HSI, a novel multi-scale weighted kernel network(MSWKNet) based on an adaptive receptive field is proposed. First, the original HSI cubic patches are transformed to the input features by combining the principal component analysis and one-dimensional spectral convolution. Then, a three-branch network with different convolution kernels is designed to convolve the input features, and adaptively adjust the size of the receptive field through the attention mechanism of each branch. Finally, the features extracted from each branch are fused together for the task of *** on three well-known hyperspectral data sets show that MSWKNet outperforms many deep learning networks in HSI classification.

关键词： multibranch structures image features convolution kernels multiple convolution kernels adaptive receptive field image classification feature extraction HSI classification single convolution kernel convolutional neural nets hyperspectral image classification representative features hyperspectral imaging geophysical image processing deep learning (artificial intelligence) multiscale weighted kernel network deep learning networks hyperspectral data sets three-branch network HSI cubic patches principal component analysis 1D spectral convolution different-sized convolution kernels

来源：评论

学校读者我要写书评

暂无评论

Indexing dynamic encrypted database in cloud for efficient secure k-nearest neighbor query

引用

Frontiers of computer science 2024年第1期18卷 187-198页

作者： Xingxin LI Youwen ZHU Rui XU Jian WANG Yushu ZHANG College of Computer Science and Technology Nanjing University of Aeronautics and AstronauticsNanjing 210016China Department of Mathematical Informatics University of TokyoTokyo 113-8654Japan Guangxi Key Laboratory of Trusted Software Guilin University of Electronic TechnologyGuilin 541004China School of Computer Science China University of GeosciencesWuhan 430074China

Secure k-Nearest Neighbor(k-NN)query aims to find k nearest data of a given query from an encrypted database in a cloud server without revealing privacy to the untrusted cloud and has wide applications in many areas,such as privacy-preservingmachine elearning gand secure biometric *** solutions have been put forward to solve this challenging ***,the existing schemes still suffer from various limitations in terms of efficiency and *** this paper,we propose a new encrypt-then-index strategy for the secure k-NN query,which can simultaneously achieve sub-linear search complexity(efficiency)and support dynamical update over the encrypted database(flexibility).Specifically,we propose a novel algorithm to transform the encrypted database and encrypted query points in the *** indexing the transformed database using spatial data structures such as the R-tree index,our strategy enables sub-linear complexity for secure k-NN queries and allows users to dynamically update the encrypted *** the best of our knowledge,the proposed strategy is the first to simultaneously provide these two *** theoretical analysis and extensive experiments,we formally prove the security and demonstrate the efficiency of our scheme.

关键词： cloud computing secure k-NN query sub-linear complexity dynamicupdate

来源：评论

学校读者我要写书评

暂无评论

Toward actionable testing of deep learning models

引用

science China(Information sciences) 2023年第7期66卷 300-302页

作者： Yingfei XIONG Yongqiang TIAN Yepang LIU Shing-Chi CHEUNG Key Laboratory of High Confidence Software Technologies(Peking University) Ministry of Education School of Computer Science Peking University Department of Computer Science and Engineering The Hong Kong University of Science and Technology School of Computer Science University of Waterloo Research Institute of Trustworthy Autonomous Systems Southern University of Science and Technology Department of Computer Science and Engineering Southern University of Science and Technology

Deep learning has become an important computational paradigm in our daily lives with a wide range of applications,from authentication using facial recognition to autonomous driving in smart vehicles. The quality of the deep learning models, i.e., neural architectures with parameters trained over a dataset, is crucial to our daily living and economy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

IoT Meets computer Vision: An Improved Detection of Tomato Pests and Diseases 17

IoT Meets Computer Vision: An Improved Detection of Tomato P...

引用

IEEE Congress on Cybermatics: 17th IEEE International Conference on Internet of Things, iThings 2024, 20th IEEE International Conference on Green Computing and Communications, GreenCom 2024, 17th IEEE International Conference on Cyber, Physical and Social Computing, CPSCom 2024, 10th IEEE International Conference on Smart Data, SmartData 2024

作者： Wang, Junling Wu, Dongyang Wang, Yupeng Nanjing University of Science and Technology ZiJin College School of Computer and Artificial Intelligence Nanjing China Nanjing Forestry University College of Information Science and Technology Nanjing China Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing China

ISBN: (纸本)9798350351637

Accurate detection of tomato pests and diseases is essential for efficient agriculture as global food demand continues to grow. Traditional methods often struggle with accurate localization and detection, highlighting the need for advanced solutions. Leveraging the Internet of Things (IoT), which facilitates interconnected devices for data acquisition and transmission, this study introduces the TPDDM algorithm to significantly enhance detection capabilities in tomato farming. To address existing challenges, we employ innovative techniques within TPDDM. Data augmentation and mosaic optimization methods are utilized to enrich the dataset and enhance image information, improving model robustness. Integration of a pre-trained Vision Transformer (ViT) into the YOLOX model boosts feature extraction and enriches the model's ability to detect subtle patterns. Additionally, the design of the Adaptive Multi-scale Feature Pyramid Network (AMFPN) enables effective fusion of features across different scales, optimizing target detection accuracy. Evaluation using comprehensive performance metrics demonstrates TPDDM's superiority over traditional approaches and other contemporary models. Real-world application of TPDDM on diverse datasets validates its high accuracy and practical utility in tomato pest and disease detection. This research underscores the transformative potential of IoT-driven technologies in advancing precision agriculture, offering scalable solutions to meet the demands of modern farming practices. © 2024 IEEE.

关键词： Fruits

来源：评论

学校读者我要写书评

暂无评论

Discovering K-Core-Truss on Multilayer Graphs 3

Discovering K-Core-Truss on Multilayer Graphs

引用

3rd International Conference on Cloud Computing, Big Data Application and Software engineering, CBASE 2024

作者： Zhuo, Licheng School of Computer Science and Engineering Nanjing University of Science and Technology Jiangsu Province Nanjing China

ISBN: (纸本)9798331506582

Multilayer graphs are an effective framework for mode ling complex systems and have garnered significant research attention. Previous studies on dense structure decomposition in multilayer graphs have generally relied on a single perspective to describe density. In contrast, this paper introduces a novel model called multilayer-k-core-truss, which combines the concepts of vertex importance as defined by both k-core and k-truss, thereby capturing relationships across multiple dimensions. Building on this dense subgraph model, we develop a basic decomposition algorithm and propose the MKCT-Index to efficiently address search problems. Extensive experiments on several real-world da-tasets validate the effectiveness and efficiency of our proposed algorithms. © 2024 IEEE.

关键词： Trusses

来源：评论

学校读者我要写书评

暂无评论

ONLINE PRIMAL DUAL MEETS ONLINE MATCHING WITH STOCHASTIC REWARDS: CONFIGURATION LP TO THE RESCUE

引用

SIAM Journal on Computing 2024年第5期53卷 1217-1256页

作者： Huang, Zhiyi Zhang, Qiankun Department of Computer Science The University of Hong Kong Hong Kong School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan China

Mehta and Panigrahi (FOCS 2012, IEEE, Piscataway, NJ, 2012, pp. 728-737) introduce the problem of online matching with stochastic rewards, where edges are associated with success probabilities and a match succeeds with the probability of the corresponding edge. It is one of the few online matching problems that have defied the randomized online primal dual framework by Devanur, Jain, and Kleinberg (SODA 2013, SIAM, Philadelphia, 2013, pp. 101-107) thus far. This paper unlocks the power of randomized online primal dual in online matching with stochastic rewards by employing the configuration linear program rather than the standard matching linear program used in previous works. Our main result is a 0.572 competitive algorithm for the case of vanishing and unequal probabilities, improving the best previous bound of 0.534 by Mehta, Waggoner, and Zadimoghaddam (SODA 2015, SIAM, Philadelphia, 2015, pp. 1388-1404) and, in fact, is even better than the best previous bound of 0.567 by Mehta and Panigrahi (FOCS 2012, IEEE, Piscataway, NJ, 2012, pp. 728-737) for the more restricted case of vanishing and equal probabilities. For vanishing and equal probabilities, we get a better competitive ratio of 0.576. Our results further generalize to the vertex-weighted case due to the intrinsic robustness of the randomized online primal dual analysis. © 2024 Society for Industrial and Applied Mathematics.

关键词： Consensus algorithm

来源：评论

学校读者我要写书评

暂无评论

MASK6D: MASKED POSE PRIORS FOR 6D OBJECT POSE ESTIMATION 49

MASK6D: MASKED POSE PRIORS FOR 6D OBJECT POSE ESTIMATION

引用

2024 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2024

作者： Xie, Yuechen Jiang, Haobo Xie, Jin PCA Lab School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China

ISBN: (纸本)9798350344851

Robust 6D object pose estimation in cluttered or occluded conditions using monocular RGB images remains a challenging task. One reason is that current pose estimation networks struggle to extract discriminative, pose-aware features using 2D feature backbones, especially when the available RGB information is limited due to target occlusion in cluttered scenes. To mitigate this, we propose a novel pose estimation-specific pre-training strategy named Mask6D. Our approach incorporates pose-aware 2D-3D correspondence maps and visible mask maps as additional modal information, which is combined with RGB images for the reconstruction-based model pre-training. Essentially, this 2D-3D correspondence maps a transformed 3D object model to 2D pixels, reflecting the pose information of the target in camera coordinate system. Meanwhile, the integrated visible mask map can effectively guide our model to disregard cluttered background information. In addition, an object-focused pre-training loss function is designed to further facilitate our network to remove the background interference. Finally, we fine-tune our pre-trained pose prior-aware network via conventional pose training strategy to realize the reliable pose prediction. Extensive experiments verify that our method outperforms previous end-to-end pose estimation methods. © 2024 IEEE.

关键词： Object Pose Estimation Prior Learning Self-Supervised Learning

来源：评论

学校读者我要写书评

暂无评论

Comprehensive optimization of piezoelectric coefficient and depolarization temperature in Mn-doped Bi_(0.5)Na_(0.5)TiO_(3)-Bi_(0.5)K_(0.5)TiO_(3)-BaTiO_(3)lead-free piezoceramics

引用

Journal of Materials science & technology 2024年第5期172卷 255-263页

作者： Huashan Zheng Enwei Sun Huajie Luo Xiaoyu Zhang Yixiao Yang Bin Yang Rui Zhang Shantao Zhang Wenwu Cao Condensed Matter Science and Technology Institute School of Instrumentation Science and EngineeringHarbin Institute of TechnologyHarbin150080China Advanced Innovation Center for Materials Genome Engineering University of Science and Technology BeijingBeijing100083China National Laboratory of Solid State Microstructures Department of Materials Science and EngineeringCollege of Engineering and Applied Science&Collaborative Innovation Center of Advanced MicrostructuresNanjing UniversityNanjing210093China Department of Mathematics and Materials Research Institute The Pennsylvania State UniversityUniversity ParkPennsylvania16802United States of America

Lead-free Bi_(_(0.5))Na_(_(0.5))TiO_(3)(BNT)piezoelectric ceramics have the advantages of large coercive fields and high Curie *** the improvement of piezoelectric coefficient(d 33)is usually accompanied by a huge sacrifice of depolarization temperature(T d).In this work,a well-balanced performance of d 33 and T d is achieved in MnO_(2)-doped 0.79(Bi_(_(0.5))Na_(_(0.5))TiO_(3))-0.14(Bi_(0.5)K_(0.5)TiO_(3))-0.07BaTiO_(3)ternary *** in-corporation of 0.25 mol%MnO_(2)enhances the d 33 by more than 40%,while T d remains almost unchanged(i.e.,d 33=181 pC/N,T d=184℃).X-ray diffraction(XRD)shows that an appropriate fraction of the small axis-ratio ferroelectric phase(pseudo-cubic,P c)coexists with the long-range ferroelectric phase(tetrag-onal,T)under this MnO_(2)*** force microscopy(PFM)has revealed a special domain configuration,namely large striped and layered macro domains embedded with small *** study provides a distinctive avenue to design BNT-based piezoelectric ceramics with high piezoelectric performance and temperature stability.

关键词： BNT-BKT-BT ceramics Mn doping Piezoelectric Depolarization temperature Phase structure

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：