检索结果-内蒙古大学图书馆

2018 IEEE Intelligent Vehicles Symposium, IV 2018

作者： Qiu, Kai Ai, Yunfeng Tian, Bin Wang, Bin Cao, Dongpu School of Aritificial Intelligence University of Chinese Academy of Sciences Beijing100049 China State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China Cloud Computing Center Chinese Academy of Sciences Dongguan523808 China School of Software Engineering University of Science and Technology of China HeFei230026 China Department of Mechanical and Mechatronics Engineering University of Waterloo WaterlooONN2L 3G1 Canada

ISBN: (纸本)9781538644522

Deep learning has made significant breakthroughs in the tasks of image classification, detection, segmentation, etc. However, the application of deep learning in robotics is still scarce. SLAM is a fundamental problem in robotics and loop closure detection is an important part of SLAM. This paper attempts to use supervised learning methods to solve the loop closure detection problem in vision SLAM. We proposed Siamese-ResNet network, which combines Siamese network with ResNet to detect loop closure. To show the effectiveness of Siamese-ResNet, we evaluate Siamese-ResNet and FabMap2.0 on several open published datasets, like TUM SLAM dataset and FabMap SLAM dataset. Compared with FabMap2.0, Siamese-ResNet shows higher accuracy, better robustness and shorter time-consuming. © 2018 IEEE.

关键词： Robotics

来源：评论

学校读者我要写书评

暂无评论

Visual Place Recognition in Long-term and Large-scale Environment based on CNN Feature

Visual Place Recognition in Long-term and Large-scale Enviro...

引用

2018 IEEE Intelligent Vehicles Symposium, IV 2018

作者： Zhu, Jianliang Ai, Yunfeng Tian, Bin Cao, Dongpu Scherer, Sebastian School of Artificial Intelligence University of Chinese Academy of Sciences Beijing100190 China State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China Cloud Computing Center Chinese Academy of Sciences Dongguan523808 China Department of Mechanical and Mechatronics Engineering University of Waterloo WaterlooONN2L 3G1 Canada Robotics Institute Carnegie Mellon University PittsburghPA15213 United States

ISBN: (纸本)9781538644522

With the universal application of camera in intelligent vehicles, visual place recognition has become a major problem in intelligent vehicle localization. The traditional solution is to make visual description of place images using hand-crafted feature for matching places, but this description method is not very good for extreme variability, especially for seasonal transformation. In this paper, we propose a new method based on convolutional neural network (CNN), by putting images into the pre-trained network model to get automatically learned image descriptors, and through some operations of pooling, fusion and binarization to optimize them, then the similarity result of place recognition is presented with the Hamming distance of the place sequence. In the experimental part, we compare our method with some state-of-the-art algorithms, FABMAP, ABLE-M and SeqSLAM, to illustrate its advantages. The experimental results show that our method based on CNN achieves better performance than other methods on the representative public datasets. © 2018 IEEE.

关键词： Hamming distance

来源：评论

学校读者我要写书评

暂无评论

M4CD: A robust change detection method for intelligent visual surveillance

arXiv

引用

arXiv 2018年

作者： Wang, Kunfeng Gou, Chao Wang, Fei-Yue State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China Innovation Center for Parallel Vision Qingdao Academy of Intelligent Industries Qingdao266000 China Research Center for Computational Experiments Parallel Systems Technology National University of Defense Technology Changsha410073 China

In this paper, we propose a robust change detection method for intelligent visual surveillance. This method, named M4CD, includes three major steps. Firstly, a sample-based background model that integrates color and texture cues is built and updated over time. Secondly, multiple heterogeneous features (including brightness variation, chromaticity variation, and texture variation) are extracted by comparing the input frame with the background model, and a multi-source learning strategy is designed to online estimate the probability distributions for both foreground and background. The three features are approximately conditionally independent, making multi-source learning feasible. Pixel-wise foreground posteriors are then estimated with Bayes rule. Finally, the Markov random field (MRF) optimization and heuristic post-processing techniques are used sequentially to improve accuracy. In particular, a two-layer MRF model is constructed to represent pixel-based and superpixel-based contextual constraints compactly. Experimental results on the CDnet dataset indicate that M4CD is robust under complex environments and ranks among the top methods. Copyright © 2018, The Authors. All rights reserved.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Parallel medical imaging for intelligent medical image analysis: Concepts, methods, and applications

arXiv

引用

arXiv 2019年

作者： Gou, Chao Shen, Tianyu Zheng, Wenbo Xue, Huadan Yu, Hui Ji, Qiang Jin, Zhengyu Wang, Fei-Yue School of Intelligent Systems Engineering Sun Yat-sen University Guangzhou510275 China The State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China School of Software Engineering Xi’an Jiaotong University Xi’an710049 China Department of Radiology Peking Union Medical College Hospital Chinese Academy of Medical Sciences Peking Union Medical College Beijing100730 China School of Creative Technologies University of Portsmouth PortsmouthPO1 2DJ United Kingdom Department of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute Qingdao Academy of Intelligent Industries Qingdao266000 China

Data-driven artificial intelligence technologies have made much progress in medical image analysis in the last decades. However, it still remains challenging due to its distinctive complexity of acquiring and annotating image data, extracting medical domain knowledge, and explaining the diagnostic decision for medical image analysis. In this paper, we propose a data-knowledge-driven framework termed as Parallel Medical Imaging (PMI) for intelligent medical image analysis based on the methodology of interactive ACP-based parallel intelligence. In the PMI framework, computational experiments with predictive learning in a data-driven way are conducted to extract medical knowledge for diagnostic decision support. Artificial imaging systems are introduced to select and prescriptively generate medical image data in a knowledge-driven way to utilize medical domain knowledge. Through the closed-loop optimization based on parallel execution, our proposed PMI framework can boost the generalization ability and alleviate the limitation of medical interpretation for diagnostic decisions. Furthermore, we illustrate the preliminary implementation of PMI method through the case studies of mammogram analysis and skin lesion image analysis. Experimental results on several public medical image datasets demonstrate the effectiveness of proposed PMI. © 2019, CC BY.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

Guide-wire Detection Using Region Proposal Network for X-ray Image-guided Navigation

Guide-wire Detection Using Region Proposal Network for X-ray...

引用

International Joint Conference on Neural Networks

作者： Li Wang Xiao-Liang Xie Gui-Bin Bian Zeng-Guang Hou Xiao-Ran Cheng Pusit Prasong State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences

ISBN: (纸本)9781509061839

Detection of surgical devices, in particular of guide-wire detection, is prerequisite during image-guided navigation in percutaneous coronary intervention (PCI). Guide-wire detection is a challenging task for following reasons: (i) X-ray images have a low signal-to-noise rate (SNR);(ii) there is a high similarity between guide-wires and some other adjacent anatomical skeletons' contours;(iii) guide-wires have various shapes and their motion is complex and nonlinear. Traditionally, guide-wires are detected using curve fitting method, and third-order B-spline curve model is always used to fit guide-wires, while B-spline fitting method has some obvious shortcomings such as it is a semi-automatic method which needs manual initialization, and it is not a real-time method because of high computational complexity. Recently, with the availability of large annotated datasets and the accessibility of hardware resources with GPUs, it is succeeded in detecting general objects with convolutional neural networks (ConvNet). In this paper, we present a novel image-based fully-automatic and real-time approach with ConvNet for guide-wires detection. ConvNet method is robust to guide-wires' various poses and other structures' effects. We evaluate our method on 22 different sequences of X-ray images. The detection accuracy evaluated by average precision (AP) reaches 89.2% and the detection speed achieves 40fps. Our experiment result shows a promising for accurate and real-time guide-wires detection in PCI navigation with ConvNet model.

关键词： navigation curve fitting method hardware resource X-rays splines (mathematics) Posing protocol control information

来源：评论

学校读者我要写书评

暂无评论

CPSS-based Signal Forwarding Method at Relays for Full-duplex Cooperative Vehicular Networks

CPSS-based Signal Forwarding Method at Relays for Full-duple...

引用

2018 IEEE Intelligent Vehicles Symposium, IV 2018

作者： Han, Shuangshuang Ma, Houxue Wang, Xiao Liu, Hong Cao, Dongpu Wang, Fei-Yue State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China Qingdao Academy of Intelligent Industries Shandong266000 China Qingdao Huituo Intelligent Machine Company Qingdao China JIANGSU XCMG CONSTRUCTION MACHINERY RESEARCH INSTITUTE LTD Xuzhou Jiangsu221004 China Mechanical and Mechatronics Engineering University of Waterloo 200 University Avenue Wes WaterlooONN2L 3G1 Canada Research Center for Military Computational Experiments and Parallel Systems Technology National University of Defense Technology Changsha410073 China

ISBN: (纸本)9781538644522

With increasing popularity of Internet of Vehicles (IoV), concerns for reliable and low complexity communication techniques are proposed due to the requirements of signal reliability and transmission delay for vehicles. Meanwhile, the explosive and pervasive use of social network applications further adds drivers' social relationships and behavioural characteristics into it, and makes it a cyber-physical-social system (CPSS). This paper proposes and analyzes an improved forward scheme for full deplex cooperative vehicular networks in terms of its CPSS features. The proposed CPSS-based forwarding (CPSS-F) strategy forwards a soft estimate of the received signal based on social historic data at the relay node (vehicle/infrastructure) to the destination node (vehicle/infrastructure), which achieves improved reliability than the two conventional strategies in cooperative networks, i.e., amplify-and-forward (AF) and decode-and-forward (DF). Furthermore, the proposed CPSS-F relay achieves performance gains and complexity reduction compared to the conventional AF and DF. Experimental results further confirm the advantages of the proposed CPSS-F for cooperative vehicular networks. The proposed CPSS-F approach is easily extended to other cooperative vehicular social networks, for example, multi-way, half-duplex, or large-antenna networks. © 2018 IEEE.

关键词： complex networks

来源：评论

学校读者我要写书评

暂无评论

Improving Auction Mechanisms for Online Real-Time Bidding Advertising with a Two-stage Resale Model

引用

IFAC-PapersOnLine 2017年第1期50卷 13575-13580页

作者： Qin, Rui Yuan, Yong Wang, Fei-Yue The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China Qingdao Academy of Intelligent Industries Qingdao China Research Center of Military Computational Experiments and Parallel System National University of Defense Technology Changsha China Beijing Engineering Research Center of Intelligent Systems and Technology Institute of Automation Chinese Academy of Sciences Beijing China

Real Time Bidding (RTB), widely publicized as one of the most promising big-data-driven business models for online computational advertising, has the potential of effectively monetizing user clicks and impressions via two-stage auctions with resale. Demand side platforms (DSPs) play a key role as intermediators in this auction process. Typically, the first-stage auctions will be conducted separately in each DSP to determine one winning advertiser registered on it, and in the second-stage auction, each DSP submits a bid based on its winning advertiser's bid to the ad exchange platform (AdX). The highest-bid DSP wins the ad impression from AdX and resells it to its winning advertiser in pursuit of the intermediate fee. This two-stage resale auction is a critical component in maintaining the effectiveness and efficiency of RTB ecosystems. In this paper, we strive to identify and design potential improvements for this auction mechanism with the aim of enhancing the total revenue of both advertisers and DSPs. We also validated our proposed auction mechanisms using the computational experiment approach. The experimental results indicate that our proposed mechanisms can make both advertisers and DSPs better off. Our work represents the first step towards a new research area of optimal mechanism design for RTB auctions, and is expected to provide useful managerial insights in RTB market practice. © 2017

关键词： Sales Commerce Machine design Auction mechanisms Computational advertisings Computational experiment Critical component Demand side Effectiveness and efficiencies Optimal mechanism designs Real time

来源：评论

学校读者我要写书评

暂无评论

Multi-view laplacian eigenmaps based on bag-of-neighbors for RGBD human emotion recognition

arXiv

引用

arXiv 2018年

作者： Liu, Shenglan Guo, Shuai Qiao, Hong Wang, Yang Wang, Bin Luo, Wenbo Zhang, Mingming Zhang, keye Du, Bixuan School of Computer Science and Technology Dalian University of Technology Dalian116024 China State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China Research Center for Brain and Cognitive Neuroscience Liaoning Normal University Dalian116024 China

Human emotion recognition is an important direction in the field of biometric and information forensics. However, most existing human emotion research are based on the single RGB view. In this paper, we introduce a RGBD video-emotion dataset and a RGBD face-emotion dataset for research. To our best knowledge, this may be the first RGBD video-emotion dataset. We propose a new supervised nonlinear multi-view laplacian eigenmaps (MvLE) approach and a multihidden-layer out-of-sample network (MHON) for RGB-D human emotion recognition. To get better representations of RGB view and depth view, MvLE is used to map the training set of both views from original space into the common subspace. As RGB view and depth view lie in different spaces, a new distance metric bag of neighbors (BON) used in MvLE can get the similar distributions of the two views. Finally, MHON is used to get the low-dimensional representations of test data and predict their labels. MvLE can deal with the cases that RGB view and depth view have different size of features, even different number of samples and classes. And our methods can be easily extended to more than two views. The experiment results indicate the effectiveness of our methods over some state-of-art methods. Copyright © 2018, The Authors. All rights reserved.

关键词： Laplace transforms

来源：评论

学校读者我要写书评

暂无评论

Three Dimensional Lenticular Display Rendering Based on Light Field Acquisition

Three Dimensional Lenticular Display Rendering Based on Ligh...

引用

SID Symposium, Seminar, and Exhibition 2017, Display Week 2017

作者： Pei, Renjing Geng, Zheng Ma, Kui Mei, Zhang Rong, Wang State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences East Zhongguanchun Road No. 95 Haidian District Beijing100190 China School of Computer and Control Engineering University of Chinese Academy of Sciences Beijing101408 China

ISBN: (纸本)9781510848757

Crosstalk is a critical defect affecting quality in 3D displays. Existing methods require tedious computations or device-specific optical measurements and results are often sub-optimal for 3D productions. We propose a method based on light field acquisition and optimization for crosstalk reduction. Algorithms were developed and experimental results showed superior performance. © 2017 SID.

关键词： Crosstalk

来源：评论

学校读者我要写书评

暂无评论

Event-triggered integral reinforcement learning for nonlinear continuous-time systems

Event-triggered integral reinforcement learning for nonlinea...

引用

IEEE Symposium Series on Computational Intelligence (SSCI)

作者： Qichao Zhang Dongbin Zhao The state Key Laboratory of Management and Control for Complex Systems University of Chinese Academy of Sciences Beijing China

In this paper, the optimal control problem for the continuous-time nonlinear systems with partially unknown dynamics is investigated. The event-triggered internal reinforcement learning (IRL) is proposed to approach the solution of the Hamilton-Jacobi-Bellman (HJB) equation. Note that the knowledge of internal dynamics is relaxed, and the event-triggered control scheme is adopted to reduce the computational burden and communication resources. For the online implementation purpose, a single-critic neural network (NN) structure is constructed to approach the optimal value function and the optimal policy with convergence analysis. Finally, a simulation example is provided to demonstrate the effectiveness of the proposed algorithm.

关键词： Artificial neural networks Heuristic algorithms Optimal control Nonlinear systems System dynamics Learning (artificial intelligence) Convergence

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：