检索结果-内蒙古大学图书馆

3rd IEEE/ACM international conference on Data Science and Advanced Analytics (DSAA)

作者： Thanh-Binh Nguyen Vu Nguyen Thuong Nguyen Venkatesh, Svetha Kumar, Mohan Dinh Phung Deakin Univ Ctr Pattern Recognit & Data Analyt Geelong Vic 3217 Australia Australian E Hlth Res Ctr CSIRO Herston Qld Australia Rochester Inst Technol Dept Comp Sci Rochester NY USA

ISBN: (纸本)9781509052066

Inferring abstract contexts and activities from heterogeneous data is vital to context-aware ubiquitous applications but still remains one of the most challenging problems. Recent advances in Bayesian nonparametric machine learning, in particular the theory of topic models based on Hierarchical Dirichlet Process (HDP), has provided an elegant solution towards these challenges. However, limited existing methods have addressed the problem of inferring latent multifaceted activities and contexts from heterogeneous data sources such as those collected from mobile devices. In this paper, we extend the original HDP to model heterogeneous data using a richer structure of the base measure being a product-space. The proposed model, called product-space HDP (PS-HDP), naturally handles the heterogeneous data from multiple sources and identify the unknown number of latent structures in a principle way. Although this framework is generic, our current work primarily focuses on inferring (latent) threefold activities of who-when-where simultaneously, which corresponds to inducing activities from data collected for identity, location and time. We demonstrate our model on synthetic data as well as on the StudentLife dataset. We provide analysis on the discovered activities and patterns to demonstrate the merit of the model. We also quantitatively evaluate the performance of PS-HDP model using standard metrics including F1-score, NMI, RI, purity, and compare them with well-known existing baseline methods.

关键词： activity recognition Bayesian nonparametrics hierarchical Dirichlet processes product-space mobile data an-alytics context-aware computing

来源：评论

学校读者我要写书评

暂无评论

Mapping the landscape of awards for research supervision: A comparison of Australia and the UK

引用

INNOVATIONS IN EDUCATION AND TEACHING international 2017年第6期54卷 601-614页

作者： Taylor, Stan McCulloch, Alistair Univ Durham Sch Educ Durham England Univ South Australia Teaching Innovat Unit Res Educ Adelaide SA Australia

The aim of this paper is to map the pattern of awards for research degree supervision in institutions in Australia and the United Kingdom. In particular, it explores the scope of such awards, their objectives, the criteria for nomination, the evidence required to be submitted, the composition of award panels, the criteria for award, and the rewards and conditions attached to success. Marked differences are found between these features in the two systems, which it is argued stem mainly from the fact that in Australia institution-led awards are the norm, while in the UK student-led awards are predominant. In conclusion, it is suggested that, while institution-led awards seem more likely to be effective in identifying, recognising and rewarding exemplary supervision, students can and should have an important role to play and, hence, that optimally all stakeholders should be involved in the awards process.

关键词： Graduate education teaching awards rewarding excellence research degree supervision doctoral education reward and recognition

来源：评论

学校读者我要写书评

暂无评论

3rd international Workshop on Modelling and Simulation for Autonomous Systems, MESAS 2016

3rd International Workshop on Modelling and Simulation for A...

引用

3rd international Workshop on Modelling and Simulation for Autonomous Systems, MESAS 2016

ISBN: (纸本)9783319476049

The proceedings contain 32 papers. The special focus in this conference is on Human Machine Integration, Interaction, Interfaces, Autonomous Systems, MS Frameworks, Architectures, Autonomous Systems Principles, Algorithms, Unmanned Aerial Vehicles, Remotely Piloted Aircraft Systems, Modelling and Simulation Application. The topics include: Challenges in representing human-robot teams in combat simulations;human-machine interface for multi-agent systems management using the descriptor function framework;a visual-haptic display for human and autonomous systems integration;modelling visual communication with UAS;using AUTOSAR high-level specifications for the synthesis of security components in automotive systems;HLA Interoperability for ROS-based autonomous systems;human driven robot grasping;an interactive framework;an open source framework for design and test of coordination protocol for asymmetric threats protection in marine environment;advancement in multi-body physics modeling for 3D graphical robot simulators;robust place recognition with combined image descriptors;geographical data and algorithms usable for decision-making process;fusion of monocular visual-inertial measurements for three dimensional pose estimation;a framework for spatio-temporal affordance maps;human-like path planning in the presence of landmarks;satcom modeling and simulation a powerful tool to enable cost effective and safe approach to RPAS Operational deployment;modelling of the UAV safety manoeuvre for the air insertion operations;sniper line-of sight calculations for route planning in asymmetric military environments;accuracy of robotic elastic object manipulation as a function of material properties and tactical decision support system to aid commanders in their decision-making.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An application program of fingerprint detection using wavelet transform for authentication 3

An application program of fingerprint detection using wavele...

引用

3rd Management and Innovation Technology international conference, MITiCON 2016

作者： Kwaochai, Aunchisa Pongyupinpanich, Surapong Areekul, Pratya San-Um, Wimol Intelligent Electronic System Research Laboratory Faculty of Engineering Thai-Nichi Institute of Technology Pattanakarn 37 Saunlaung Bangkok10250 Thailand Master Program of Forensic Informatics Ramkhamhang University Bangkapi Bangkok10240 Thailand Department of Mathematics Computer Science Chulachomklao Royal Military Academy Province Nakornnayok26001 Thailand

ISBN: (纸本)9781509041053

This paper presents the application program of fingerprint detection using wavelet transform for authentication. Fingerprints are obtained from the site of crime, old documents and excavated things. This paper proposes a fingerprint recognition technique based on wavelet-based texture pattern recognition method. By collecting incomplete fingerprints of 2 people with 6 images each. The original seven of uncompleted fingerprint images were undetected and only one was detected. After enhancing image quality using wavelet transform method, it was found that those of seven images were completed. The comparison method was a matching technique. This work focuses on the ability of the program and knowledge of image processing by studying algorithms of image processing. In design algorithm process, different technique enhancements were employed as well as the study of different of filters and algorithms based on existing research results from literature. © 2016 IEEE.

关键词： Wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

Collaborator Effort Optimisation in Multimodal Crowdsourcing for Transcribing Historical Manuscripts 3rd

Collaborator Effort Optimisation in Multimodal Crowdsourcing...

引用

5th Iberian SLTech Workshop / 3rd international conference on Advances in Speech and Language Technologies for Iberian Languages (IberSPEECH)

作者： Granell, Emilio Martinez-Hinarejos, Carlos-D. Univ Politecn Valencia Pattern Recognit & Human Language Technol Res Ctr Camino Vera S-N E-46022 Valencia Spain

ISBN: (纸本)9783319491691;9783319491684

Crowdsourcing is a powerful tool for massive transcription at a relatively low cost, since the transcription effort is distributed into a set of collaborators, and therefore, supervision effort of professional transcribers may be dramatically reduced. Nevertheless, collaborators are a scarce resource, which makes optimisation very important in order to get the maximum benefit from their efforts. In this work, the optimisation of the work load in the side of collaborators is studied in a multimodal crowdsourcing platform where speech dictation of handwritten text lines is used as transcription source. The experiments explore how this optimisation allows to obtain similar results reducing the number of collaborators and the number of text lines that they have to read.

关键词： Crowdsourcing framework Collaborator effort Speech recognition Document transcription

来源：评论

学校读者我要写书评

暂无评论

23rd international conference on Neural Information Processing, ICONIP 2016

23rd International Conference on Neural Information Processi...

引用

23rd international conference on Neural Information Processing, ICONIP 2016

ISBN: (纸本)9783319466712

The proceedings contain 296 papers. The special focus in this conference is on Machine Learning, Neuromorphic Hardware, Sensory Perception, pattern recognition, Social Networks, Brain-Machine Interface and Computer Vision. The topics include: Non-parametric e-mixture of density functions;an entropy estimator based on polynomial regression with Poisson error structure;a theoretical analysis of semi-supervised learning;on the noise resilience of ranking measures;group dropout inspired by ensemble learning;semi supervised autoencoder;modal regression via direct log-density derivative estimation;active consensus-based semi-supervised growing neural gas;nuclear norm regularized randomized neural network;data cleaning using complementary fuzzy support vector machine technique;character-aware convolutional neural networks for paraphrase identification;hyper-parameter tuning for graph kernels via multiple kernel learning;hinge loss projection for classification;acceleration of Word2vec Using GPUs;group information-based dimensionality reduction via canonical correlation analysis;pattern and frequency generation using an opto-electronic reservoir computer with output feedback;a neural network model for retaining object information required in a categorization task;a hybrid pooling method for convolutional neural networks;a tag probability correlation based microblog recommendation method;sentiment and behavior analysis of one controversial American individual on twitter;continuous vigilance estimation using LSTM neural networks;assessing the properties of single-trial fixation-related potentials in a complex choice task;content-based image retrieval using deep search;detection of human faces using neural networks and energy-based multi-plane detection from 3D point clouds.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Document analysis for music scores via machine learning 2016

Document analysis for music scores via machine learning

引用

3rd international Workshop on Digital Libraries for Musicology, DLfM 2016

作者： Calvo-Zaragoza, Jorge Vigliensoni, Gabriel Fujinaga, Ichiro Pattern Recognition and Artificial Intelligence Group Department of Software and Computing Systems University of Alicante Alicante Spain Centre for Interdisciplinary Research in Music Media and Technology Schulich School of Music McGill University MontréalQC Canada

ISBN: (纸本)9781450347518

Content within musical documents not only contains musical notation but can also include text, ornaments, annotations, and editorial data. Before any attempt at automatic recognition of elements in these layers, it is necessary to perform a document analysis process to detect and classify each of its constituent parts. The obstacle for this analysis is the high heterogeneity amongst collections, which makes it difficult to propose methods that can be generalizable to a broader range of sources. In this paper we propose a data-driven document analysis framework based on machine learning, which focuses on classifying regions of interest at pixel level. The main advantage of this approach is that it can be exploited regardless of the type of document provided, as long as training data is available. Our preliminary experimentation includes a set of specific tasks that can be performed on music such as the detection of staff lines, isolation of music symbols, and the layering of the document into its elemental parts. © 2016 ACM.

关键词： Music

来源：评论

学校读者我要写书评

暂无评论

3rd iapr TC3 Workshop on Multimodal pattern recognition of Social Signals in Human-Computer-Interaction, MPRSS 2014

引用

3rd iapr TC3 Workshop on Multimodal pattern recognition of Social Signals in Human-Computer-Interaction, MPRSS 2014

ISBN: (纸本)9783319148984

The proceedings contain 18 papers. The special focus in this conference is on pattern recognition, Machine Learning, and Information Fusion Methods with Applications in Social Signal Processing. The topics include: Automatic image collection of objects with similar function by learning human grasping forms;client specific image gradient orientation for unimodal and multimodal face representation;multiple-manifolds discriminant analysis for facial expression recognition from local patches set;noisy speech recognition based on combined audio-visual classifiers;complementary Gaussian mixture models for multimodal speech recognition;fusion of text and audio semantic representations through CCA;a real time gesture recognition system for human computer interaction;a SIFT-based feature level fusion of iris and ear biometrics;audio-visual user identification in HCI scenarios;online smart face morphing engine with prior constraints and local geometry preservation and exploring alternate modalities for tag recommendation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Evaluation of face recognition using vector features in local pattern descriptors

Evaluation of face recognition using vector features in loca...

引用

international conference on Devices, Circuits and Systems (ICDCS)

作者： S. Valarmathy M. Arun Kumar R. Sangeetha Department of ECE Bannari Amman Institute of Technology

ISBN: (纸本)9781509023103

The local feature descriptors have received more attention due to the effectiveness in the field of face recognition. The existing LTrP method works based on the first order derivatives along horizontal and vertical directions. This LVP method demonstrates the relation between the center pixel and its neighbor pixels. The feature vector gets increased and the recognition rate gets reduced due to the noise present in the higher order derivatives. But the local vector pattern effectively extracts more detailed discriminative information in a given sub-region than the other local pattern descriptors. CST used in the vector pattern can suppress the slight noise influence in the images. Furthermore, K-NN has been used as a classifier in this approach. Three well-known databases such as ORL, Yale and JAFFE face database are used in the performance evaluation. The experimental result clearly shows that the LVP method gives us a better performance and analyzed with the variation of number of neighbors pixels.

关键词： Face recognition Feature extraction Databases Lighting Data mining Circuits and systems Face

来源：评论

学校读者我要写书评

暂无评论

A pattern recognition approach for identification of transducer-structure debonding using Lamb waves

A pattern recognition approach for identification of transdu...

引用

international conference on Devices, Circuits and Systems (ICDCS)

作者： P Durai Arun C. N. Sathyanarayana S. Raja V P S Naidu National Aerospace Laboratories CSIR Bangalore Karnataka IN CSIR-National Aerospace Laboratories Bangalore India

ISBN: (纸本)9781509023103

In structural health monitoring, using piezoelectric transducers to generate high frequency elastic waves like Lamb waves in the structures is eminent. In general, piezoelectric transducers are assumed to be perfectly bonded with the host structure; however, in practical environment, there are possibilities for them to have faults. Since detecting, locating and assessing damages in a structure depend solely on transducer responses, transducer fault identification is vital. By using electrical admittance, axial strain and shear stress as function of frequencies or in analog interface circuits to identify faulty transducers, lead to demand of circuitry and processes, consequently increasing the implementation complexity. Hence, we propose a pattern recognition system that can identify transducers that are partially bonded to host structure. This pattern recognition system employs classification for features extracted from instantaneous Lamb wave signals with no need of baseline data.

关键词： Transducers Circuit faults Feature extraction Continuous wavelet transforms Fault diagnosis

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：