检索结果-内蒙古大学图书馆

Integrating Dense LiDAR-Camera Road Detection Maps by a Multi-Modal CRF Model

IEEE Transactions on Vehicular Technology 2019年第12期68卷 11635-11645页

作者： Gu, Shuo Zhang, Yigong Tang, Jinhui Yang, Jian Alvarez, Jose M. Kong, Hui PCA Lab Key Lab of Intelligent Percept. and Syst. for High-Dimensional Information of Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security Sch. of Comp. Sci. and Engineering Nanjing Univ. of Sci. and Technol. Nanjing 210094 China School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing 210094 China

Road detection is an important task in autonomous navigation systems. In this paper, we propose a road detection method via a LiDAR-camera fusion strategy to exploit both the range and color information. The whole system consists of three parts. In the LiDAR based part, we transform the discrete 3D LiDAR point clouds to continuous 2D LiDAR range images and propose a distance-aware height-difference based scanning approach to get the road estimations quickly. In the camera based part, we apply a light-weight transfer learning based road segmentation network. In the LiDAR-camera fusion part, we transform the detection results from LiDAR and camera to dense and binary ones to solve the data imbalance problem and fuse them in a multi-modal conditional random field (MM-CRF) framework. Experiments show that the proposed MM-CRF fusion method can operate in real-time and achieve competitive performance compared with the state-of-the-art road detection algorithms on the KITTI-Road benchmark. © 1967-2012 IEEE.

关键词： KITTI LiDAR-camera fusion road detection

来源：评论

学校读者我要写书评

暂无评论

A probability weighted beamformer for noise robust ASR 19

A probability weighted beamformer for noise robust ASR

引用

19th Annual Conference of the International Speech Communication, INTERSPEECH 2018

作者： Bu, Suliang Zhao, Yunxin Hwang, Mei-Yuh Sun, Sining Dept. of Electrical Engineering and Computer Science University of Missouri Columbia United States Mobvoi AI Lab Redmond WA United States Sch. of Computer Science Northwestern Polytechnical University Xi'an China

We investigate a novel approach to spatial filtering that is adaptive to conditions at different time-frequency (TF) points for noise removal by taking advantage of speech sparsity. Our approach combines a noise reduction beamformer with a minimum variance distortionless response (MVDR) beamformer or Generalized Eigenvalue (GEV) beamformer through TF posterior probabilities of speech presence (PPSP). To estimate PPSP, we study both statistical model-based and neural network based methods, where in the former, we use complex Gaussian mixture modeling (CGMM) on temporally augmented spatial spectral features, and in the latter, we use neural network (NN) based TF masks to initialize speech and noise covariance matrices in CGMM. We have conducted experiments on CHiME-3 task. On its real noisy speech test set, our methods of feature augmentation, TF dependent spatial filter, and NN-based mask initialization on covariances for CGMM have yielded relative word error rate (WER) reductions cumulatively by 8%, 16%, and 25% over the original CGMM based MVDR. On the real test data, the three methods have also produced consistent WER reductions when replacing MVDR by GEV. © 2018 International Speech Communication Association. All rights reserved.

关键词： Noise abatement

来源：评论

学校读者我要写书评

暂无评论

A cooperative sch.duling framework for shared transportation services 19

A cooperative scheduling framework for shared transportation...

引用

19th COTA International Conference of Transportation Professionals: Transportation in China - Connecting the World, CICTP 2019

作者： Wang, Xinli Qiao, Yanan Wang, Tianyu Guo, Jifu Zhou, Xuesong Xian, Kai Tong, Lu Carol Xie, Jindong Natl. Eng. Lab. for Compreh. Transp. Big Data Applic. Technol. Sch. of Transp. Sci. and Engineering Beihang Univ. Beijing100191 China State Key Lab of Software Development Environment School of Computer Science and Engineering Beihang Univ. Beijing100191 China Beijing Transport Institute No. 9 LiuLiQiao South Lane Fengtai District China

ISBN: (数字)9780784482292

ISBN: (纸本)9780784482292

To improve the efficiency of metropolitan shared transportation services (STS), this paper proposes a cooperative sch.duling framework to avoid congestion. Two procedures are considered: shared path sch.duling and congestion solving. In a known static traffic network, classic vehicle routing problem (VRP) solutions may obtain an optimal path. But with an increasing number of shared transportation services required, congestion is likely, triggered by open loop VRP sch.duling, especially when shared vehicles overwhelm private ones. Dynamic traffic assignment (DTA) solutions could be employed for short-term traffic situation prediction. However, VRP and DTA cannot be combined linearly, due to uncontrolled recursive calls of routing procedures within VRP and DTA. Hence, a software structure is further studied to guarantee convergence of the cooperative framework. Using two open-source transportation packages, VRPLite and DTALite, we test our framework with the Western 3rd Ring Road of the Beijing regional network. Results verify the framework's usability and flexibility. © ASCE.

关键词： sch.duling

来源：评论

学校读者我要写书评

暂无评论

A Robust Nonlinear Microphone Array Postfilter for Noise Reduction

A Robust Nonlinear Microphone Array Postfilter for Noise Red...

引用

International Workshop on Acoustic Signal Enhancement (IWAENC)

作者： Suliang Bu Yunxin Zhao Mei-Yuh Hwang Sining Sun Dept. of Electrical Engineering and Computer Science University of Missouri-Columbia USA Mobvoi AI Lab Redmond WA USA Sch. of Computer Science Northwestern Polytechnical University Xi'an China

We propose a robust nonlinear microphone array postfilter for noise reduction. This postfilter is formulated as a function of noise power ratio before and after beamforming and a local speech-to-observation power ratio. The two ratios are readily obtained during beamforming, and can be approximated by local speech posterior probability or time-frequency masks of neural network. This avoids the difficulty in estimating local speech and noise variances of a beamformed signal. On the CHiME-3 test set, we have evaluated our proposed postfilter in comparison with two other postfiltering methods, and our proposed postfilter has produced the best objective scores on the simulated noisy speech as well as higher listening preference scores on real noisy speech.

关键词： Array signal processing Noise measurement Estimation Microphone arrays Noise reduction

来源：评论

学校读者我要写书评

暂无评论

Resilience or Robustness: Identifying Topological Vulnerabilities in Rail Networks

arXiv

引用

arXiv 2018年

作者： Pagani, Alessio Mosquera, Guillem Alturki, Aseel Johnson, Samuel Jarvis, Stephen Wilson, Alan Guo, Weisi Varga, Liz Alan Turing Institute United Kingdom Mathematics Inst. University of Warwick United Kingdom Dept. of Computer Science University of Warwick United Kingdom Sch. of Mathematics University of Birmingham United Kingdom Sch. of Engineering University of Warwick United Kingdom Sch. of Management Cranfield University United Kingdom

Many critical infrastructure systems have network structure and are under stress. Despite their national importance, the complexity of large-scale transport networks means we do not fully understand their vulnerabilities to cascade failures. The research in this paper examines the interdependent rail networks in Greater London and surrounding commuter area. We focus on the morning commuter hours, where the system is under the most demand stress. There is increasing evidence that the topological shape of the network plays an important role in dynamic cascades. Here, we examine whether the different topological measures of resilience (stability) or robustness (failure) are more appropriate for understanding poor railway performance. The results show that resilience and not robustness has a strong correlation to the consumer experience statistics. Our results are a way of describing the complexity of cascade dynamics on networks without the involvement of detailed agent-based-models, showing that cascade effects are more responsible for poor performance than failures. The network science analysis hints at pathways towards making the network structure more resilient by reducing feedback loops. Copyright © 2018, The Authors. All rights reserved.

关键词： Autonomous agents

来源：评论

学校读者我要写书评

暂无评论

Towards a Personalized Item Recommendation Approach in Social Tagging Systems Using Intuitionistic Fuzzy DBSCAN

Towards a Personalized Item Recommendation Approach in Socia...

引用

International Conference on Intelligent Human-Machine Systems and Cybernetics, IHMSC

作者： Chun Guan Yuen Kevin Kam Fung Yong Yue Department of Computer Science and Software Engineering Xi'an Jiaotong-Liverpool University Suzhou China Sch. of Bus. Univ. of Social Sci. Singapore Singapore

In folksonomies, users annotate items with abundant personalized tags. The tags can be used in recommendation systems to produce meaningful information. The Density-Based Spatial Clustering of Applications with Noise (DBSCAN) can be applied in recommendation systems. This paper proposes the Intuitionistic Fuzzy DBSCAN (IF-DBSCAN) for the personalized item recommendation in folksonomy. The IF-DBSCAN is used to cluster items with respect to the user-defined tags. The Intuitionistic Fuzzy Set (IFS) is used to represent tag values which are vague and uncertain. DBSCAN can cluster items with the tags represented by using IFSs into different groups. An example of movie recommendation is demonstrated for the applicability of the proposed method.

关键词： Motion pictures Clustering algorithms Fuzzy sets Tagging computer science Software engineering Euclidean distance

来源：评论

学校读者我要写书评

暂无评论

Covert Visuospatial Attention (VSA) for EEG-Based Asynchronous Control of Robot

Covert Visuospatial Attention (VSA) for EEG-Based Asynchrono...

引用

IEEE International Conference on Systems, Man and Cybernetics

作者： Oh Yoke Chew Neethu Robinson K.G. Smitha School of Computer Science and Engineering Nanyang Technological University Singapore Sch. of Comput. Sci. & Eng. Nanyang Technological University Singapore Singapore SG

Brain-computer Interface (BCI) is an effective modality for direct communication between brain and computer, bypassing brain's conventional communication pathway of the nerves and muscles. The objective of this work is to demonstrate EEG-based asynchronous control to a robotic device (Sphero) using covert visuospatial attention (VSA). Alpha and beta power band features in conjunction with a forward search technique for feature selection are used as input features for training a classifier model. The selected features are used to train Linear Discriminant Analysis (LDA) model. Average validation accuracy of 71.25% and average test accuracy of 66 % are obtained from eight subjects. We also investigated the effect of inter-session variations in feature patterns and accuracy of control using a multiple online VSA session. It is observed that there is a shift of discriminative features within subject across sessions. An asynchronous control session is also done and average test accuracy of 62.7% is reported which validates our objective for the work..

关键词： Training Robots Rhythm Feature extraction Visualization Task analysis Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Development of Navigation System in Field Robot for Forest Management

Development of Navigation System in Field Robot for Forest M...

引用

International Symposium on Soft Computing and Intelligent Systems (SCIS)

作者： Ayumu Tominaga Hayashi Eiji Abbe Mowshowitz Grad. Sch. of Information Engineering Kyushu Institute of Technology Izuka Fukuoka Japan Kyushu Kogyo Daigaku Kitakyushu Fukuoka JP Depertment of Computer Science The City College of New York New York NY USA

ISBN: (纸本)9781538626344;9781538626337

This paper addresses the problem of using a mobile, autonomous robot to manage a forest whose trees are destined for eventual harvesting. This was accomplished by 1) computing a trajectory for the robot in advance of its entrance into the forest, and 2) developing a program and equipping the robot with the instruments needed to follow the trajectory. Computation of a trajectory in a forest is facilitated by treating the trees as vertices in a graph. Current, laser-based instruments make it possible to identify individual trees and compute distances between them. With this information, a forest can be represented as a weighted graph. This graph can then be modified systematically in a way that allows for computing a Hamiltonian circuit that passes between each pair of trees. Our robot "SOMA" built on an ATV platform, was able to follow a part of the trajectory computed for this small forest, thus demonstrating the feasibility of forest maintenance by an autonomous, labor-saving robot.

关键词： Forestry Vegetation Collision avoidance Trajectory Robot kinematics Service robots

来源：评论

学校读者我要写书评

暂无评论

Dart: Domain-adversarial residual-transfer networks for unsupervised cross-domain image classification

arXiv

引用

arXiv 2018年

作者： Fang, Xianghong Bai, Haoli Guo, Ziyi Shen, Bin Hoi, Steven Xu, Zenglin SMILE Lab Univ. of Electronic Sci and Tech.of China Chengdu Sichuan611731 China Dept Computer Science and Engineering Chinese University of Hong Kong Shatin N. T. Hong Kong SMILE Lab Sch. of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu Sichuan611731 China Google Inc. Singapore Management University

The accuracy of deep learning (e.g., convolutional neural networks) for an image classification task critically relies on the amount of labeled training data. Aiming to solve an image classification task on a new domain that lacks labeled data but gains access to cheaply available unlabeled data, unsupervised domain adaptation is a promising technique to boost the performance without incurring extra labeling cost, by assuming images from different domains share some invariant characteristics. In this paper, we propose a new unsupervised domain adaptation method named Domain-Adversarial Residual-Transfer (DART) learning of Deep Neural Networks to tackle cross-domain image classification tasks. In contrast to the existing unsupervised domain adaption approaches, the proposed DART not only learns domain-invariant features via adversarial training, but also achieves robust domainadaptive classification via a residual-transfer strategy, all in an end-to-end training framework. We evaluate the performance of the proposed method for cross-domain image classification tasks on several well-known benchmark data sets, in which our method clearly outperforms the state-of-the-art approaches. Copyright © 2018, The Authors. All rights reserved.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Detection of Breast Tumour Tissue Regions in Histopathological Images using Convolutional Neural Networks

Detection of Breast Tumour Tissue Regions in Histopathologic...

引用

International Image Processing, Applications and Systems Conference (IPAS)

作者： Yibao Sun Zhaoyang Xu Carina Strell Carlos Fernández Moro Fredrik Wärnberg Le Dong Qianni Zhang EECS QMUL London UK Dept. of Oncology-Pathology Karolinska Institute Stockholm Sweden Dept. of Clinical Pathology/Cytology Karolinska University Hospital Stockholm Sweden Dept. of Surgical Sciences Uppsala Academic Hospital Uppsala Sweden Sch. of Computer Science and Engineering Univ. of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9781728102481;9781728102474

Ductal carcinoma in situ (DCIS) is considered a pre-invasive breast cancer and sometimes it can develop into an invasive ductal carcinoma. The analysis of histopathological images to detect tumour border of DCIS could provide important information for better diagnosis of patients. We present a deep learning based system to automatically identify DCIS in histopathological images. Specifically, a convolutional neural network (CNN) is first trained to predict labels of small patches cropped out of a histopathological whole slide image. Next, a sliding window method is used to produce a probability map of DCIS. Finally, given the probability map, a tumor border of DCIS is produced and delineated with the method of Marching Cubes to facilitate pathologists' review and assessment. Evaluation of cross validation demonstrates that the CNN model of GoogleNet performs well in histology image patch classification with an overall accuracy of (98.46±0.40)% and identifies the DCIS tissue patches with a F1-score of (97.40±1.18)% (mean±variance). Moreover, around 95.6% tumour tissue within the enclosed tumour regions can be identified by our developed method. Finally, the goal of tumor border detection can be well achieved with a few post-processing steps.

关键词： Tumors Training Image segmentation Data mining Feature extraction Conferences

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：