检索结果-内蒙古大学图书馆

Object Detection Algorithm for Improving Non-Maximum Suppression Using GIoU

学校读者我要写书评

暂无评论

Object Detection Algorithm for Improving Non-Maximum Suppres...

作者： Zhiqiang Hou Xiaoyi Liu Lilin Chen College of Computer Science and Technology Xian University of Posts and Telecommunications Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi'an University of Posts and Telecommunications

To address the problem of object miss detection and object false detection in single threshold-Non-Maximum Suppression algorithm,this paper proposed a GDT-NMS(Generalized Intersection over Union Dual Threshold NMS,GDT-NMS)) algorithm which using GIoU(Generalized Intersection over Union).Using the GIoU indicator computing the similarity between objects,can better describe the relative position and overlap between the *** we proposed the dual-threshold NMS algorithm,which can balance the relationship between the object missed detection problem and the object false detection problem,reduce "false positive example" *** nonlinearly processing the weight function,the object is better *** algorithm uses Faster R-CNN as the *** experimental results show that the improved algorithm has outstanding performance.

关键词： GDT NMS Object Detection Algorithm for Improving Non-Maximum Suppression Using GIoU

An improved scan matching algorithm in SLAM

学校读者我要写书评

暂无评论

An improved scan matching algorithm in SLAM

International Conference on Systems and Informatics (ICSAI)

作者： Hongkai Zhang Niansheng Chen Guangyu Fan Dingyu Yang Shanghai DianJi University School of Electronic Information Engineering Shanghai China Alibaba Group Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou China

ISBN: (数字)9781728152561

ISBN: (纸本)9781728152578

Simultaneous localization and mapping (SLAM) technology has always been the research focus of robot navigation in unknown environment. Aiming at the problem of cumulative errors of robot pose in the localization process of SLAM algorithm based on particle filter, a loop detection algorithm based on graph-SLAM was proposed. The algorithm uses constraints to adjust the robot attitude at different moments. In this paper, the constraint refers to the scanning matching of lidar. In the process of drawing, when the robot returns to the known area, if the current laser scanning is successfully matched with the previous laser scanning, the robot's posture can be adjusted to eliminate the accumulated errors caused by the odometer. In the process of laser scanning matching, the method of grouping step threshold value judgment is proposed to match the laser point cloud, which can effectively reduce the computation. Experimental results show that the proposed algorithm can effectively eliminate the cumulative errors of positioning and achieve a better mapping effect.

关键词： Three-dimensional displays Lasers Jacobian matrices Simultaneous localization and mapping Mathematical model Laser radar

Hyperspectral image Classification with Spatial Consistence Using Fully Convolutional Spatial Propagation Network

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Jiang, Yenan Li, Ying Zou, Shanrong Zhang, Haokui Bai, Yunpeng School of Computer Science National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Shaanxi Provincial Key Laboratory of Speech and Image Information Processing Northwestern Polytechnical University Xi’an710129 China School of Computer Science University of Adelaide AdelaideSA5005 Australia National Key Laboratory of Science and Technology on Space Microwave Xi’an710000 China School of Computing and Information Systems University of Melbourne MelbourneVIC3010 Australia

In recent years, deep convolutional neural networks (CNNs) have demonstrated impressive ability to represent hyperspectral images (HSIs) and achieved encouraging results in HSI classification. However, the existing CNN-based models operate at the patch-level, in which pixel is separately classified into classes using a patch of images around it. This patch-level classification will lead to a large number of repeated calculations, and it is hard to identify the appropriate patch size that is beneficial to classification accuracy. In addition, the conventional CNN models operate convolutions with local receptive fields, which cause the failure of contextual spatial information modeling. To overcome these aforementioned limitations, we propose a novel end-to-end, pixel-to-pixel fully convolutional spatial propagation network (FCSPN) for HSI classification. Our FCSPN consists of a 3D fully convolution network (3D-FCN) and a convolutional spatial propagation network (CSPN). Specifically, the 3D-FCN is firstly introduced for reliable preliminary classification, in which a novel dual separable residual (DSR) unit is proposed to effectively capture spectral and spatial information simultaneously with fewer parameters. Moreover, the channel-wise attention mechanism is adapted in the 3D-FCN to grasp the most informative channels from redundant channel information. Finally, the CSPN is introduced to capture the spatial correlations of HSI via learning a local linear spatial propagation, which allows maintaining the HSI spatial consistency and further refining the classification results. Experimental results on three HSI benchmark datasets demonstrate that the proposed FCSPN achieves state-of-the-art performance on HSI classification. Copyright © 2020, The Authors. All rights reserved.

关键词： Convolution

DSRGAN: Explicitly learning disentangled representation of underlying structure and rendering for image generation without tuple supervision

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Hao, Guang-Yuan Yu, Hong-Xing Zheng, Wei-Shi School of Data and Computer Science Sun Yat-sen University Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education Collaborative Innovation Center of High Performance Computing Nudt Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou

We focus on explicitly learning disentangled representation for natural image generation, where the underlying spatial structure and the rendering on the structure can be independently controlled respectively, yet using no tuple supervision. The setting is significant since tuple supervision is costly and sometimes even unavailable. However, the task is highly unconstrained and thus ill-posed. To address this problem, we propose to introduce an auxiliary domain which shares a common underlying-structure space with the target domain, and we make a partially shared latent space assumption. The key idea is to encourage the partially shared latent variable to represent the similar underlying spatial structures in both domains, while the two domain-specific latent variables will be unavoidably arranged to present renderings of two domains respectively. This is achieved by designing two parallel generative networks with a common Progressive Rendering Architecture (PRA), which constrains both generative networks' behaviors to model shared underlying structure and to model spatially dependent relation between rendering and underlying structure. Thus, we propose DSRGAN (GANs for Disentangling Underlying Structure and Rendering) to instantiate our method. We also propose a quantitative criterion (the Normalized Disentanglability) to quantify disentanglability. Comparison to the state-of-The-Art methods shows that DSRGAN can significantly outperform them in disentanglability. Copyright © 2019, The Authors. All rights reserved.

关键词： Rendering (computer graphics)

An improved index based on mapreduce for path queries in public transportation networks 14

学校读者我要写书评

暂无评论

An improved index based on mapreduce for path queries in pub...

14th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery, ICNC-FSKD 2018

作者： Ning, Zhiqing Dai, Genan Liu, Yubao Ge, Youming Wu, Jiefeng School of Data and Computer Science Sun Yat-sun University Guangzhou510006 China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou510006 China

ISBN: (纸本)9781538680971

In this paper, we consider the path queries in public transportation networks which are widely used in path planning and location-based services. In general, path index is often adopted to speed up the path queries in public transportation networks. However, the pre-processing for building path index for the existing best-known query algorithms is often not efficient. Aimed at this shortcoming, we propose an improved index, which is based on MapReduce for path queries. Specifically, the proposed algorithm includes the map and reduce phases. In map phase, the public transportation network is divided into different subgraphs and generate key-value pairs of index for subgraphs. In reduce phase, the result of map function are aggregated by subgraphs' key-value pairs to obtain the index. The experimental results based on the real datasets show the effectiveness of the proposed algorithms. © 2018 IEEE.

关键词： MapReduce

Sparse generalized canonical correlation analysis: Distributed alternating iteration based approach

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Cai, Jia Lv, Kexin Huo, Junyi Huang, Xiaolin Yang, Jie School of Statistics and Mathematics Guangdong University of Finance & Economics Big Data and Educational Statistics Application Laboratory 21 Chisha Road Guangzhou Guangdong510320 China Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University MOE Key Laboratory of System Control and Information Processing 800 Dongchuan Road Shanghai200240 China School of Electronics and Computer Science University of Southampton University Road SouthamptonSO17 1BJ United Kingdom

Sparse canonical correlation analysis (CCA) is a useful statistical tool to detect latent information with sparse structures. However, sparse CCA works only for two datasets, i.e., there are only two views or two distinct objects. To overcome this limitation, in this paper, we propose a sparse generalized canonical correlation analysis (GCCA), which could detect the latent relations of multiview data with sparse structures. Moreover, the introduced sparsity could be considered as Laplace prior on the canonical variates. Specifically, we convert the GCCA into a linear system of equations and impose 1 minimization penalty for sparsity pursuit. This results in a nonconvex problem on Stiefel manifold, which is difficult to solve. Motivated by Boyd’s consensus problem, an algorithm based on distributed alternating iteration approach is developed and theoretical consistency analysis is investigated elaborately under mild conditions. Experiments on several synthetic and real world datasets demonstrate the effectiveness of the proposed algorithm. Copyright © 2020, The Authors. All rights reserved.

关键词： Statistical mechanics

Multiple residual dense networks for reconfigurable intelligent surfaces cascaded channel estimation

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Jin, Yu Zhang, Jiayi Huang, Chongwen Yang, Liang Xiao, Huahua Ai, Bo Wang, Zhiqin The School of Electronic and Information Engineering Beijing Jiaotong University Beijing100044 China The Frontiers Science Center for Smart High-speed Railway System Beijing Jiaotong University Beijing100044 China Zhejiang Provincial Key Lab of Information Processing Communication and Networking Zhejiang University Hangzhou310007 China College of Computer Science and Electronic Engineering Hunan University Changsha410082 China ZTE Corporation State Key Laboratory of Mobile Network Mobile Multimedia Technology Shenzhen518057 China State Key Laboratory of Rail Traffic Control and Safety Beijing Jiaotong University Beijing100044 China Henan Joint International Research Laboratory of Intelligent Networking and Data Analysis Zhengzhou University Zhengzhou450001 China Research Center of Networks and Communications Peng Cheng Laboratory Shenzhen518055 China China Academy of Information and Communications Technology Beijing100191 China

Reconfigurable intelligent surface (RIS) constitutes an essential and promising paradigm that relies programmable wireless environment and provides capability for space-intensive communications, due to the use of low-cost massive reflecting elements over the entire surfaces of man-made structures. However, accurate channel estimation is a fundamental technical prerequisite to achieve the huge performance gains from RIS. By leveraging the low rank structure of RIS channels, three practical residual neural networks, named convolutional blind denoising network, convolutional denoising generative adversarial networks and multiple residual dense network, are proposed to obtain accurate channel state information, which can reflect the impact of different methods on the estimation performance. Simulation results reveal the evolution direction of these three methods and reveal their superior performance compared with existing benchmark schemes. Copyright © 2021, The Authors. All rights reserved.

关键词： Channel estimation

A deep neural information fusion architecture for textual network embeddings

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Xu, Zenan Su, Qinliang Quan, Xiaojun Zhang, Weijia School of Data and Computer Science Sun Yat-Sen University Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou China School of Electronics and Information Technology Sun Yat-Sen University

Textual network embeddings aim to learn a low-dimensional representation for every node in the network so that both the structural and textual information from the networks can be well preserved in the representations. Traditionally, the structural and textual embeddings were learned by models that rarely take the mutual influences between them into account. In this paper, a deep neural architecture is proposed to effectively fuse the two kinds of informations into one representation. The novelties of the proposed architecture are manifested in the aspects of a newly defined objective function, the complementary information fusion method for structural and textual features, and the mutual gate mechanism for textual feature extraction. Experimental results show that the proposed model outperforms the comparing methods on all three datasets. Copyright © 2019, The Authors. All rights reserved.

关键词： Network embeddings

Parallelized Top-k Route Search with User's Preferences

学校读者我要写书评

暂无评论

Parallelized Top-k Route Search with User's Preferences

IEEE International Conference on Big data and Cloud Computing (BdCloud)

作者： Sen Zhang Youming Ge Qun Jiang Haihang Li Yubao Liu School of Data and Computer Science Sun Yat-Sen University Guangzhou P.R.China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou P.R.China

ISBN: (数字)9781728143286

ISBN: (纸本)9781728143293

In this paper, we consider the top-k route search with user's preferences. Specifically, given a set of POIs, our problem is to find k different routes from a source POI to a target POI such that the constraint on route cost and the POIs covered by the route can optimally satisfy the user-defined weighted feature preference. It has been shown that the problem is NP-hard. The challenge is how to select from plenty of POIs and construct an optimal route especially when the size of candidate POIs is large. In order to support top-k route search on a large dataset or with a looser budget constraint, we propose a parallel method on a single machine to speed up the search. Moreover, we adopt further effective pruning strategies to reduce the search space. The experimental results on real-world datasets show that our proposed parallel method is much efficient, about 10 times faster than the existing serial algorithm at best.

关键词： Roads Search problems Parallel algorithms Big data Task analysis Aggregates