检索结果-内蒙古大学图书馆

5th international workshop on Computational Color Imaging, CCIW 2015

ISBN: (纸本)9783319159782

the proceedings contain 23 papers. the special focus in this conference is on Color reproduction, Color sensation and perception, Color image processing, Spectral imaging and Color in digital cultural heritage. the topics include: Color and spectral mixings in printed surfaces;an overview of color name applications in computer vision;optics and computational methods for hybrid resolution spectral imaging;relative pose estimation and fusion of 2D spectral and 3D lidar images;representation of cultural objects by image sets with directional illumination;descriptor-based image colorization and regularization;experimental analysis for modeling color of halftone images;development of a dynamic relighting system for moving planar objects with unknown reflectance;collection, analysis and representation of memory color information;the display gamut available to simulate colors perceived by anomalous trichromats;visual search for normal color and dichromatic observers using a unique distracter color;first stage of a human visual system simulator;the color logarithmic image processing (CoLIP) antagonist space and chromaticity diagram;does color influence image complexity perception?;adaptive filters for color images: median filtering and its extensions;can linear data projection improve hyperspectral face recognition?;spectral impulse noise model for spectral image processing;evaluation of the colorimetric performance of single-sensor image acquisition systems employing colour and multispectral filter array;high-resolution multispectral scanning for mesoscopic investigation of discoloration of traditional Japanese pigments and estimating the colors of paintings.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Knowledge graph Completion Model Combining graph Convolutional Networks with Self-Attention Mechanisms

A Knowledge Graph Completion Model Combining Graph Convoluti...

引用

pattern recognition and Machine Learning (PRML), IEEE international Conference on

作者： Yao Chen Shaorong Li Hao Fang Yaokun Li Jianhong Hu Yunnan Transportation Science Research Institute Co. Ltd. Kunming China School of Software Yunnan University Kunming China

ISBN: (数字)9798350355925

ISBN: (纸本)9798350355932

Knowledge graph completion aims to predict missing entities or relationships in a knowledge graph, addressing the issue of data sparsity. In recent years, deep learning-based models, especially graph Convolutional Network (GCN) and Self-Attention mechanisms, have gained attention for their advantages in handling graph-structured data. We propose a knowledge graph completion model, CSANN, which integrates GCN and self-attention mechanisms. First, GCN module is used to learn the initial representations of entities and relationships, considering the neighboring information of entities. Next, the initial representations of entities and relationships are concatenated to form a sequence, which is further refined by introducing the self-attention mechanism. By combining the local graph structure awareness of GCN and the sequential dependency capturing ability of self-attention mechanisms, CSANN improves the accuracy of knowledge graph completion. Experiments conducted on two public datasets, FB15K-237 and WN18RR, show that the proposed model outperforms many classical models, such as TransE, DistMult, ConvE, and RotatE, in terms of Mean Reciprocal Rank (MRR) and Hit@K.

关键词： Measurement Knowledge engineering Adaptation models graph convolutional networks Computational modeling Scalability Knowledge graphs Machine learning Magnetic heads pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics): Preface

Lecture Notes in Computer Science (including subseries Lectu...

引用

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2013年 7877 LNCS卷 VI页

作者： Kropatsch, Walter G. Artner, Nicole M. Haxhimusa, Yll Jiang, Xiaoyi Vienna University of Technology Department of Pattern Recognition and Image Processing Favoritenstr. 9-11 186-3 1040 Vienna Austria University of Münster Department of Computer Science Einsteinstr. 62 48149 Münster Germany

来源：评论

学校读者我要写书评

暂无评论

Generating visualization-based analysis scenarios from maintenance task descriptions

Generating visualization-based analysis scenarios from maint...

引用

IEEE international workshop on Visualizing Software for Understanding and Analysis (VISSOFT)

作者： Salima Hassaine Karim Dhambri Houari Sahraoui Pierre Poulin DIRO University of Montreal Canada

Software visualization is an efficient and flexible tool to inspect and analyze software data at various levels of detail. However, software analysts typically do not have a sufficient background in visualization and cognitive science to select efficient representations and parameters without the help of visualization experts. To overcome this problem, we propose an approach to generate software analysis tasks that use visualization. To this end, we use taxonomies of low-level analytic tasks, high-level interactive tasks, and perceptual rules to design an assistant that proposes analysis scenarios.

关键词： Data visualization Software tools Taxonomy Data analysis Independent component analysis Software maintenance Software quality Cognitive science pattern recognition Humans

来源：评论

学校读者我要写书评

暂无评论

2-D object recognition by structured neural networks in a pyramidal architecture

Computer Architectures for Machine Perception, Proceedings (...

引用

Computer Architectures for Machine Perception, Proceedings (CAMP) 2000年 81-86页

作者： Cantoni, Virginio Petrosino, Alfredo Universita di Pavia Pavia Italy

In the paper we propose an approach to the realization of models inspired to biological solutions for pattern recognition. the approach is based on a hierarchical modular structure capable to learn by examples and recognize objects in digital images. the adopted techniques are based on multiresolution image correlation and neural networks. Performance on two different data sets and experimental timings on a SIMD machine are also reported.

关键词： pattern recognition

来源：评论

学校读者我要写书评

暂无评论

5th IEEE/ACIS international Conference on Computer and Information Science and 1st IEEE/ACIS international workshop on Component-based Software Engineering,Software Architecture and Reuse - Title

5th IEEE/ACIS International Conference on Computer and Infor...

引用

international Conference on Computer and Information Science (ACIS)

the following topics are dealt with: parallel and distributed computing; software metrics and project management; communication systems and networks; data mining; data warehousing; information management systems; Internet; mobile computing; wireless computing; software engineering; information engineering; management information systems; image processing and pattern recognition; computer architecture and software testing; artificial intelligence; intelligent agent technology; Web engineering

关键词：

来源：评论

学校读者我要写书评

暂无评论

Hierarchical interactive image segmentation using irregular pyramids

引用

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2011年 6658 LNCS卷 245-254页

作者： Gerstmayer, Michael Haxhimusa, Yll Kropatsch, Walter G. Pattern Recognition and Image Processing Group Institute of Computer Graphics and Algorithms Vienna University of Technology Austria

ISBN: (纸本)9783642208430

In this paper we describe modifications of irregular image segmentation pyramids based on user-interaction. We first build a hierarchy of segmentations by the minimum spanning tree based method, then regions from different (granularity) levels are combined to a final (better) segmentation with user-specified operations guiding the segmentation process. based on these operations the users can produce a final image segmentation that best suits their applications. this work can be used for applications where we need accuracy in image segmentation, in annotating images or creating ground truth among others. © 2011 Springer-Verlag Berlin Heidelberg.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Generalized planted (l,d)-motif problem with negative set

引用

5th international workshop on Algorithms in Bioinformatics, WABI 2005

作者： Leung, Henry C. M. Chin, Francis Y. L. Department of Computer Science University of Hong Kong Pofulam Hong Kong

ISBN: (纸本)3540290087

Finding similar patterns (motifs) in a set of sequences is an important problem in Computational Molecular Biology. Pevzner and Sze [18] defined the planted (l,d)-motif problem as trying to find a length-l pattern that occurs in each input sequence with at most d substitutions. When d is large, this problem is difficult to solve because the input sequences do not contain enough information on the motif. In this paper, we propose a generalized planted (l,d)-motif problem which considers as input an additional set of sequences without any substring similar to the motif (negative set) as extra information. We analyze the effects of this negative set on the finding of motifs, and define a set of unsolvable problems and another set of most difficult problems, known as "challenging generalized problems". We develop an algorithm called VANS based on voting and other novel techniques, which can solve the (9,3), (11,4),(15,6) and (20,8)-motif problems which were unsolvable before as well as challenging problems of the planted (l,d)-motif problem such as (9,2), (11,3), (15,5) and (20,7)-motif problems. © Springer-Verlag Berlin Heidelberg 2005.

关键词： pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Dual graph contraction for irregular pyramids

Dual graph contraction for irregular pyramids

引用

international Conference on pattern recognition

作者： D. Willersinn V.G. Kropatsch Department for Pattern Recognition and Image Processing Technical University of of Vienna Wien Austria

A new algorithm for building irregular pyramids is presented. the algorithm is based on only two basic operations on graphs, contraction and removal of edges. By making use of the concept of dual graphs, the algorithm overcomes the problem of unbounded vertex degree proper to the existing approach to building irregular pyramids. this boundedness extends the scope of parallel, degree preserving graph contraction also to irregular structures. Our method can be applied to all problems in which a 2D discrete space can be represented by a planar graph. Four-connected square grids, region adjacency graphs, Voronoi diagrams and Delaunay triangulations are examples of such plane representations.

关键词： Stochastic processes Image segmentation Automation pattern recognition Image processing Buildings Robustness Clustering algorithms Computational efficiency Process control

来源：评论

学校读者我要写书评

暂无评论

Semantic Bottlenecks: Quantifying and Improving Inspectability of Deep representations 1

引用

42nd German Conference on pattern recognition, DAGM GCPR 2020 held in parallel with 25th international Symposium on Vision, Modeling, and Visualization, VMV 2020 and 10th Eurographics workshop on Visual Computing for Biology and Medicine, VCBM 2020

作者： Losch, Max Maria Fritz, Mario Schiele, Bernt Max Planck Institute for Informatics Saarland Informatics Campus Saarbrücken Germany CISPA Helmholtz Center for Information Security Saarland Informatics Campus Saarbrücken Germany

ISBN: (数字)9783030712785

ISBN: (纸本)9783030712778

Today’s deep learning systems deliver high performance based on end-to-end training but are notoriously hard to inspect. We argue that there are at least two reasons making inspectability challenging: (i) representations are distributed across hundreds of channels and (ii) a unifying metric quantifying inspectability is lacking. In this paper, we address both issues by proposing Semantic Bottlenecks (SB), integrated into pretrained networks, to align channel outputs with individual visual concepts and introduce the model agnostic AUiC metric to measure the alignment. We present a case study on semantic segmentation to demonstrate that SBs improve the AUiC up to four-fold over regular network outputs. We explore two types of SB-layers in this work: while concept-supervised SB-layers (SSB) offer the greatest inspectability, we show that the second type, unsupervised SBs (USB), can match the SSBs by producing one-hot encodings. Importantly, for both SB types, we can recover state of the art segmentation performance despite a drastic dimensionality reduction from 1000s of non aligned channels to 10s of semantics-aligned channels that all downstream results are based on. © 2021, Springer Nature Switzerland AG.

关键词： Semantics

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：