检索结果-内蒙古大学图书馆

World Congress on Intelligent control and Automation

作者： Leijie Zhang Zhiqiang Cao Xiangrui Meng Chao Zhou Shuo Wang State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing 100190 China

ISBN: (纸本)9781467384155

Depth map provides rich information and it can be utilized in object tracking to handle some challenging problems in conventional RGB tracking such as occlusions and model drift. In this paper, we present a tracker that provides an effective real-time target tracking method based on a binocular camera. The proposed tracker is an extension of the popular KCF algorithm that leverages a circulant structure of tracking-by-detection with kernels for tracking. On this basis, we design a simple yet effective method to detect occlusions and recover tracking using noisy depth map obtained from a binocular camera. Firstly, one needs to identify exact and reasonable peak number of the target region's depth histogram and apply GMM model to evaluate the depth distribution. The occlusion is then detected based on the depth evaluation and the maximum response from KCF. When the occlusion happens, it is segmented to build the corresponding search region for recovery. The experimental results demonstrate the effectiveness of the proposed method and the superiority to the KCF tracker.

关键词： Target tracking Kernel Cameras Matrices Algorithm design and analysis Histograms Correlation

来源：评论

学校读者我要写书评

暂无评论

Convolutional Fitted Q Iteration For Vision-Based control Problems

Convolutional Fitted Q Iteration For Vision-Based Control Pr...

引用

International Joint Conference on Neural Networks

作者： Dongbin Zhao Yuanheng Zhu Le Lv Yaran Chen Qichao Zhang State Key Laboratory of Management and Control for Complex Systems Institution of Automation Chinese Academy of Sciences Beijing 100190 China

ISBN: (纸本)9781509006212

In this paper a deep reinforcement learning (DRL) method is proposed to solve the control problem which takes raw image pixels as input states. A convolutional neural network (CNN) is used to approximate Q functions, termed as Q-CNN. A pretrained network, which is the result of a classification challenge on a vast set of natural images, initializes the parameters of Q-CNN. Such initialization assigns Q-CNN with the features of image representation, so it is more concentrated on the control tasks. The weights are tuned under the scheme of fitted Q iteration (FQI), which is an offline reinforcement learning method with the stable convergence property. To demonstrate the performance, a modified Food-Poison problem is simulated. The agent determines its movements based on its forward view. In the end the algorithm successfully learns a satisfied policy which has better performance than the results of previous researches.

关键词： Feature extraction Artificial neural networks Learning (artificial intelligence) Biological neural networks Image representation Convergence Aerospace electronics

来源：评论

学校读者我要写书评

暂无评论

A Robot Pose Estimation Approach Based on key Feature Registration

A Robot Pose Estimation Approach Based on Key Feature Regist...

引用

World Congress on Intelligent control and Automation

作者： Wenbo Yuan Tianzhu Wang Zhiqiang Cao Min Tan State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing 100190 China

ISBN: (纸本)9781467384155

Pose estimation is a fundamental issue in the research of mobile robot. A robot pose estimation approach based on key feature registration is proposed in this paper. The key features in the environment are extracted firstly. When dealing with many non-corresponding points, registration has a slow convergence rate, and gains poor performance. To solve these problems, a key Feature Based Accelerated Iterative Closest Point (KF-AICP) algorithm is presented for key feature registration. Then the current robot pose can be estimated based on the result of registration. The dataset is established using 2D laser radar in indoor scene. The experimental results show that the proposed approach achieves good accuracy and real-time in pose estimation of mobile robot. The validity of proposed approach is verified.

关键词： Iterative closest point algorithm Pose estimation Laser radar Feature extraction Mobile robots Data models

来源：评论

学校读者我要写书评

暂无评论

A Novel Model Analysis Method and Dynamic Modelling for Hybrid Structure Flexible Manipulator

A Novel Model Analysis Method and Dynamic Modelling for Hybr...

引用

IEEE International Conference on Mechatronics and Automation

作者： Teng Long En Li Guodong Yang Zize Liang The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing 100190 China

ISBN: (纸本)9781509023974

Tip deflection and vibration have a large impact on the manipulator operation safety, and hybrid structure flexible manipulator has a larger workspace and more flexibility. In this paper, a new modeling scheme for a hybrid structure flexible manipulator is proposed. In the proposed scheme, the joint variables of hybrid structure flexible manipulator are divided in rigid joint angle and small joint angle deviations. The prismatic joint is divided into three segments according to the structure features of flexible manipulator. The three segments are the foundation segment, overlap segment and extension segment. Each segment is modeled by the theory of the Euler-Bernoulli beam, and the elastic deflection is described by assumed mode method, the kinetic energy and potential energy of every segment is analyzed respectively. Base on the previous researches of modeling for multi-link flexible manipulator, the dynamic equations of the manipulator are obtained by using the Lagrangian approach. In order to verify the correctness of model and analyze the characteristics of model, the comparative simulation experiments are designed. Tip vibration, angle deviation, vibration frequency and amplitude are observed, and the characteristics of the model are analyzed according to the experiment results.

关键词： model analysis dynamic modelling hybrid structure flexible manipulator

来源：评论

学校读者我要写书评

暂无评论

Motion detection based on improved Sobel and ViBe algorithm

Motion detection based on improved Sobel and ViBe algorithm

引用

Chinese control Conference (CCC)

作者： Yujia Zhang Xiaoguang Zhao Min Tan State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing Beijing CN

Motion detection plays a crucial role in most video based applications. A particular background subtraction technique called ViBe (Visual Background Extractor) is commonly used to obtain foreground objects from the background due to its high detection rate and low computational complexity. However, the performance is not very satisfying. Therefore, this paper presents an improved ViBe algorithm to increase the accuracy and robustness of motion detection. Specifically, a foreground feature map is created by optimizing the result of ViBe algorithm. Then the edge detection of the original video frames is achieved after pre-sharpening using improved Sobel operator and Otsu algorithm. Finally, by feature fusion (of the foreground and background feature maps) and contour filling, the motion detection results can be obtained. The experiments demonstrate the improvements of the proposed modifications at a limited additional cost.

关键词： Image edge detection Motion detection Feature extraction Filtering algorithms Signal processing algorithms Maximum likelihood detection Nonlinear filters

来源：评论

学校读者我要写书评

暂无评论

The paralleleye dataset: Constructing large-scale artificial scenes for traffic vision research

arXiv

引用

arXiv 2017年

作者： Li, Xuan Wang, Kunfeng Tian, Yonglin Yan, Lan Wang, Fei-Yue School of Automation Beijing Institute of Technology Beijing100081 China State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China Qingdao Academy of Intelligent Industries Qingdao266000 China Department of Automation University of Science and Technology of China Hefei230027 China Research Center for Computational Experiments and Parallel Systems Technology National University of Defense Technology Changsha410073 China

Video image datasets are playing an essential role in design and evaluation of traffic vision algorithms. Nevertheless, a longstanding inconvenience concerning image datasets is that manually collecting and annotating large-scale diversified datasets from real scenes is time-consuming and prone to error. For that virtual datasets have begun to function as a proxy of real datasets. In this paper, we propose to construct large-scale artificial scenes for traffic vision research and generate a new virtual dataset called "ParallelEye". First of all, the street map data is used to build 3D scene model of Zhongguancun Area, Beijing. Then, the computer graphics, virtual reality, and rule modeling technologies are utilized to synthesize large-scale, realistic virtual urban traffic scenes, in which the fidelity and geography match the real world well. Furthermore, the Unity3D platform is used to render the artificial scenes and generate accurate ground-truth labels, e.g., semantic/instance segmentation, object bounding box, object tracking, optical flow, and depth. The environmental conditions in artificial scenes can be controlled completely. As a result, we present a viable implementation pipeline for constructing large-scale artificial scenes for traffic vision research. The experimental results demonstrate that this pipeline is able to generate photorealistic virtual datasets with low modeling time and high accuracy labeling. Copyright © 2017, The Authors. All rights reserved.

关键词： Pipelines

来源：评论

学校读者我要写书评

暂无评论

Competitive perspective identification via topic based refinement for online documents

Competitive perspective identification via topic based refin...

引用

IEEE International Conference on Intelligence and Security Informatics, ISI

作者： Junjie Lin Wenji Mao Daniel Zeng State Key Laboratory of Management and Control for Complex Systems Chinese Academy of Sciences School of Computer and Control Engineering University of Chinese Academy of Sciences

ISBN: (纸本)9781509038664

People write online documents from different personal perspectives. The competitive perspectives they hold reflect the conflicts in their fundamental stances and viewpoints. For many security-related applications, it is both beneficial and critical to identify the competitive perspectives implied in online documents. Previous work on competitive perspective identification is based on word features, which did not consider that the word usage for perspective expression varies with topics in documents. Thus topic information can be incorporated and contribute to a more fine-grained treatment of perspective identification. Motivated by this, this paper proposes an approach for competitive perspective identification in online documents via topic based refinement. Our approach refines the basic word feature-based perspective identification model with latent semantic information. In addition, we develop a self-adaptive process to fit the model parameters automatically. Experimental study shows the effectiveness of our approach compared to the related work and the baseline methods.

关键词： Mathematical model Training Support vector machines Semantics Fitting Optimization Testing

来源：评论

学校读者我要写书评

暂无评论

Labelling Topics in Weibo Using Word Embedding and Graph-Based Method

Labelling Topics in Weibo Using Word Embedding and Graph-Bas...

引用

International Conference on Information systems Engineering (ICISE)

作者： Zhipeng Jin Qiudan Li Can Wang Daniel D. Zeng Lei Wang The State Key Laboratory of Management and Control for Complex Systems Chinese Academy of Sciences Beijing China Department of Management Information Systems University of Arizona Tucson Arizona USA

Nowadays, in China, Weibo is becoming an increasingly popular way for people to know what is happening in the world. Labelling topics is of much importance for better understanding the semantics of topics. Existing works mainly focus on deriving candidate labels by exploring the use of external knowledge, which may be more appropriate for well formatted and static documents. Recently, it has been a new trend to generate labels for sparse and dynamic microblogging environment using summarization method. The challenges of labelling topics are how to obtain coherent candidate labels and how to rank the labels. In this paper, based on the latest research work in deep learning, we propose a novel and unified model for labelling topics in Weibo, which firstly adopts word embedding and clustering method to learn dense semantic representation of topic words and mine the coherent candidate topic labels, then, generates interpretable labels using a graph-based model. Experimental results show that topics labels discovered by our model not only have high topic coherence, but also are meaningful and interpretable.

关键词： Labeling Semantics Coherence Machine learning Xenon Modeling Heuristic algorithms

来源：评论

学校读者我要写书评

暂无评论

Robust Traffic Sign Detection in complex Road Environments

Robust Traffic Sign Detection in Complex Road Environments

引用

2016年IEEE汽车电子与安全国际会议

作者： Bin Tian Ran Chen Yanjie Yao Naiqiang Li the State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Qingdao Academy of Intelligent Industries

Traffic sign recognition is an important topic in driver assistant system and intelligent autonomous vehicles. Traffic sign detection is a critical step, whose performance greatly affect the performance and computation cost of traffic sign recognition. In this paper, we propose a traffic sign detection method based on a scoring SVM model. First, traffic sign color and color gradient are extracted according to their color characteristics. Then, the shape of traffic sign is computed by a voting scheme, yielding shape score maps. After that, the score maps of traffic signs are used to train a SVM model. Compared with single traffic sign score value, the score map is more efficient to verify the existence of a traffic sign. Finally, the trained SVM model is used to detect traffic signs. Experiments show that the proposed method is more effective than the voting based method to detect traffic signs.

关键词： image Robust Traffic Sign Detection in complex Road Environments

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation for Multi-antenna Vehicular Communication Based on IEEE 802.11p Standard

Performance Evaluation for Multi-antenna Vehicular Communica...

引用

International Conference on Computing, Networking and Communications

作者： Shuangshuang Han Kai Zhao Liu-Qing Yang Xiang Cheng The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Qingdao Academy of Intelligent Industries

ISBN: (纸本)9781467385800

Multi-input multi-output (MIMO) has been a promising technique for wireless communication areas, such as intelligent transportation systems. In this paper, we present performance evaluation of a multi-antenna transceiver for Wireless Access in Vehicular Environments (WAVE) according to IEEE 802.11p standard. Herein, MIMO WAVE is performed by employing three MIMO schemes, namely, Alamouti, vertical BLAST (V-BLAST) and space shift keying (SSK) scheme. The above three MIMO schemes are discussed and compared, considering the influence of different driving velocity and with/without line of sight (LOS) on the system performance. Based on the analysis and simulation results, it can be found that compared with single antenna systems, V-BLAST scheme would significantly increase the transmission data rate, while Alamouti and SSK scheme could improve the reliability and robustness even under poor receive conditions.

关键词： MIMO 802.11p Alamouti V-BLAST SSK

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：