检索结果-内蒙古大学图书馆

2nd Indian International Conference on Artificial Intelligence, IICAI 2005

作者： Nanda, P.K. Patra, D. Pradhan, A. Department of Electrical Engineering Image Processing and Computer Vision Laboratory National Institute of Technology Rourkela Orissa 769008 India

ISBN: (纸本)0972741216

In this paper, we propose a hybrid Tabu Expectation Maximization (TEM) Algorithm for segmentation of Brain Magnetic Resonance (MR) images in both supervised and unsupervised framewrok. Gaussian Hidden Markov Random Field (GHMRF) is used to model the available degraded image. In supervised framework, the apriori image MRF model parameters as well as the GHMRF model parameters are assumed to be known. The class labels are estimated using the Maximum a Posteriori (MAP) estimation criterion. In unsupervised framework, the problem of model parameter estimation and label estimation is formulated in Expectation Maximization (EM) framework. The labels are estimated using the proposed Tabu Search algorithm while the model parameters are the maximum likelihood estimates. Our proposed algorithm yields results with arbitrary initial paramater set and thus overcomes the problem of proper choice of initial parameters. The results obtained are comparable with the results obtained by using the algorithm proposed by Zhang ***. [15] , where the Iterated Conditional Mode (ICJM) algorithm is used for computing the MAP estimates. Copyright © IICAI 2005.

关键词： Tabu search

来源：评论

学校读者我要写书评

暂无评论

IMPROVED SPEECH READING THROUGH A FREE-PARTS REPRESENTATION

IMPROVED SPEECH READING THROUGH A FREE-PARTS REPRESENTATION

引用

2005 International Conference on Auditory-Visual Speech processing, AVSP 2005

作者： Lucey, Simon Lucey, Patrick Advanced Multimedia Processing Laboratory Department of Electrical and Computer Engineering Carnegie Mellon University PittsburghPA15213 United States Speech Audio Image and Video Research Laboratory Queensland University of Technology GPO Box 2424 Brisbane4001 Australia

Motivated by the success of free-parts based representations in face recognition [1] we have attempted to address some of the problems associated with applying such a philosophy to the task of speaker-independent automatic speech reading. Hitherto, a major problem with canonical area-based approaches in automatic speech reading is the intrinsic lack of training observations due to the visual speech modality’s low sample rate and large variability in appearance. We believe a free-parts representation can overcome many of these limitations due to its natural ability to generalize by producing many observations from a single mouth image, whilst still preserving the ability to discriminate between various visual-speech units. This approach additionally requires a modification to traditional techniques employed for the estimation of hidden Markov Models (HMMs), whose resultant models we currently refer to as free-parts HMMs (FP-HMMs). Results will be presented on the CUAVE audiovisual speech database. © AVSP 2005. All rights reserved.

关键词： Hidden Markov models

来源：评论

学校读者我要写书评

暂无评论

Using a free-parts representation for visual speech recognition

Using a free-parts representation for visual speech recognit...

引用

Digital Imaging Computing: Techniques and Applications, DICTA 2005

作者： Lucey, Patrick Lucey, Simon Sridharan, Sridha Speech Audio Image and Video Research Laboratory Queensland University of Technology GPO Box 2424 Brisbane 4001 Australia Advanced Multimedia Processing Laboratory Department of Electrical and Computer Engineering Carnegie Mellon University Pittsburgh PA 15213 United States

ISBN: (纸本)0769524672

Motivated by the success of free-parts based representations in face recognition, we have attempted to address some of the problems associated with applying such a philosophy to the task of speaker-independent visual speech recognition. A major problem with canonical area-based approaches in automatic visual speech recognition is the dependence these approaches have on locating and tracking the speaker's region of interest (ROI) correctly. By employing a free-parts representation, we assume that the position/structure of patches within the mouth image can be relaxed so they can "freely" move to varying extents, hence reducing the influence of the front-end effect. In this paper, we show that by using a free-parts representation we gain some robustness against the problem of ROI localisation and tracking compared to current area-based feature extraction techniques such as the discrete cosine transform (DCT). Also in this paper, we expose the importance of representation for the task of visual speech recognition highlighted by the poor results current representations yield. © 2005 IEEE.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

220 fs Er-Yb: Glass laser mode-locked by a broadband low-loss Si/Ge saturable absorber

220 fs Er-Yb: Glass laser mode-locked by a broadband low-los...

引用

Advanced Solid-State Photonics, ASSP 2005

作者： Grawert, F.J. Gopinath, J.T. Ilday, F.O. Shen, H.M. Ippen, E.P. Kärtner, F.X. S. Akiyama Liu, J. Wada, K. Kimerling, L.C. Research Laboratory of Electronics Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology 77 Massachusetts Avenue Cambridge MA 02139 United States Material Processing Center Department of Materials Science and Engineering Massachusetts Institute of Technology 77 Massachusetts Avenue Cambridge MA 02139 United States

We report a silicon/germanium saturable Bragg reflector, compatible with CMOS processing. Its sub-picosecond recovery time enables a C-band spanning mode-locked Er-Yb:glass laser at 1540 nm, generating 220 fs pulses. ... 详细信息

ISBN: (纸本)1557527814

关键词： Glass

来源：评论

学校读者我要写书评

暂无评论

Ontology-Based Access to Multimedia Cultural Heritage Collections - The REACH Project

Ontology-Based Access to Multimedia Cultural Heritage Collec...

引用

Eurocon

作者： C. Doulaverakis Y. Kompatsiaris M.G. Strintzis Centre for Research and Technology Informatics and Telematics Institute Thessaloniki Greece Information Processing Laboratory Electrical and Computer Engineering Department Aristotle University of Thessaloniki Thessaloniki Greece

In this paper, an approach for providing unified access to heterogeneous distributed cultural heritage collections is presented. The proposed approach is developed within the framework of the REACH Greek National Project with the overall aim to provide a Web portal enabling access to different databases of cultural content with emphasis on multimedia material. In order to achieve this, the project employs ontology-based representation of cultural content, multimedia content-based search algorithms, hybrid ontology and content-based access and user-friendly search interfaces. This fusion approach provides users with resources for building queries of multimedia repositories sequentially using multiple individual search tools, and performs higher quality searching. In the current stage of development, real content such as inscriptions and coins from the Greco-Roman time period, has successfully been incorporated in the system

关键词： Ontologies Cultural differences Portals Multimedia databases Information retrieval Content based retrieval Material storage Multimedia systems Informatics Telematics

来源：评论

学校读者我要写书评

暂无评论

A genetic algorithm-based approach to knowledge-assisted video analysis

A genetic algorithm-based approach to knowledge-assisted vid...

引用

IEEE International Conference on Image processing

作者： N. Voisine S. Dasiopoulou F. Precioso V. Mezaris I. Kompatsiaris M.G. Strintzis Informatics and Telematics Institute Centre for Research and Technology Hellas Thessaloniki Greece Information Processing Laboratory Electrical and Computer Engineering Department Aristotle University of Thessaloniki Thessaloniki Greece

Efficient video content management and exploitation requires extraction of the underlying semantics, a non-trivial task associating low-level features of the image domain and high-level semantic descriptions. In this paper, a knowledge-assisted approach for extracting semantics of domain-specific video content is presented. Domain knowledge considers both low-level features (color, motion, shape) and spatial behavior (topological and directional information). During the preprocessing step, a set of over-segmented homogenous atom-regions is generated and their low-level and spatial descriptions are extracted. A genetic algorithm is then applied in order to find the optimal interpretation according to a specific domain conceptualization. The proposed approach was tested on the formula one, tennis and beach vacations domains showing promising results.

关键词： Algorithm design and analysis Data mining Ontologies Genetic algorithms Object detection Content management Feature extraction Knowledge representation Semantic Web Laboratories

来源：评论

学校读者我要写书评

暂无评论

About room response equalization and dereverberation

About room response equalization and dereverberation

引用

IEEE Workshop on Applications of Signal processing to Audio and Acoustics

作者： M. Karjalainen T. Paatero J.N. Mourjopoulos P.D. Hatziantoniou Laboratory of Acoustics and Audio Signal Processing Helsinki University of Technology Finland Audio Group Wire Communications Lab Electrical and Computer Engineering Department University of Patras Patras Greece

Improvement of sound reproduction in rooms by signal processing techniques has been studied seriously for more than twenty years. Such methods are called for example dereverberation, room (response) equalization, and room (response) correction. Understanding the limitations and quality criteria for such methods is gradually achieved through studies of this highly multidimensional problem. In the present paper we discuss the terminology and goals, compare some recent approaches of these techniques in practical cases, and finally point out future challenges

关键词： Acoustic signal processing Multidimensional signal processing Multidimensional systems Terminology

来源：评论

学校读者我要写书评

暂无评论

Using a Free-Parts Representation for Visual Speech Recognition

Using a Free-Parts Representation for Visual Speech Recognit...

引用

Proceedings of the Digital Image Computing: Technqiues and Applications (DICTA)

作者： P. Lucey S. Lucey S. Sridharan Queensland University of Technology Speech Audio Image and Video Research Laboratory Brisbane Australia Advanced Multimedia Processing Laboratory Department of Electrical and Computer Engineering Carnegie Mellon University Pittsburgh PA USA

Motivated by the success of free-parts based representations in face recognition, we have attempted to address some of the problems associated with applying such a philosophy to the task of speaker-independent visual speech recognition. A major problem with canonical area-based approaches in automatic visual speech recognition is the dependence these approaches have on locating and tracking the speaker’s region of interest (ROI) correctly. By employing a free-parts representation,we assume that the position/structure of patches within the mouth image can be relaxed so they can "freely" move to varying extents, hence reducing the influence of the front-end effect. In this paper, we show that by using a free-parts representation we gain some robustness against the problem of ROI localisation and tracking compared to current area-based feature extraction techniques such as the discrete cosine transform (DCT). Also in this paper, we expose the importance of representation for the task of visual speech recognition highlighted by the poor results current representations yield.

关键词： Speech recognition Mouth Hidden Markov models Laboratories Face recognition Feature extraction Discrete cosine transforms Robustness Automatic speech recognition Pixel

来源：评论

学校读者我要写书评

暂无评论

Effective memory reduction scheme used for reference frames in H.264/AVC video codec

Effective memory reduction scheme used for reference frames ...

引用

Midwest Symposium on Circuits and Systems (MWSCAS)

作者： Chih-Chang Chen Sheng-Shiung Chen O.T.C. Chen Chien-Fa Chen Fang-Chu Chen Department of Electrical Engineering Signal and Media Laboratories National Chung Cheng University Taiwan Digital Signal Processing Tech. Department of Computer & Communications Research Laboratory Industrial Technology and Research Institute Taiwan

In the H.264/AVC video codec, multiple reference frames are applied for motion estimation to provide good coding efficiency. However, multiple reference frames also require more memory cost than one reference frame used in a video codec. To minimize the amount of memory requirement, this work proposes to add a decision unit in the H.264/AVC codec to store these reference frames. The said decision unit would look closely at macroblock (MB) of the currently decoded frame to determine it's storage type to be compressed or reconstructed data, according to the encoding parameters of subblocks in this MB. Furthermore, an adaptive scheme is proposed to store the currently decoded frame as reconstructed picture for lowering the computational complexity and using the buffer in full. Experimental results reveal that the proposed memory reduction scheme requires lower complexity and better picture quality than the conventional work. Therefore, the proposed memory reduction scheme can be widely used in various applications of the H.264/AVC codec to minimize the memory requirement.

关键词： Automatic voltage control Video codecs Decoding Buffer storage Motion estimation Costs Computational complexity Videoconference Video compression Encoding

来源：评论

学校读者我要写书评

暂无评论

A computer-aided system for mass detection and classification in digitized mammograms

引用

Biomedical engineering - Applications, Basis and Communications 2005年第5期17卷 215-228页

作者： Yang, Sheng-Chih Wang, Chuin-Mu Chung, Yi-Nung Hsu, Giu-Cheng Lee, San-Kan Chung, Pau-Choo Chang, Chein-I. Department of Electrical Engineering National Cheng Kung University Tainan Taiwan Department of Computer Science and Information Engineering National Chin Yi Institute of Technology Taichung Taiwan Department of Electrical Engineering Da-Yeh University Chunghua Taiwan Department of Radiology Tri-Service General Hospital National Defense Medical Center Taipei Taiwan Administrative Office Suao Veterans Hospital Yilan Taiwan Remote Sensing Signal and Image Processing Laboratory Department of Computer Science and Electrical Engineering University of Maryland Baltimore MD United States

This paper presents a computer-assisted diagnostic system for mass detection and classification, which performs mass detection on regions of interest followed by the benign-malignant classification on detected masses. In order for mass detection to be effective, a sequence of preprocessing steps are designed to enhance the intensity of a region of interest, remove the noise effects and locate suspicious masses using five texture features generated from the spatial gray level difference matrix (SGLDM) and fractal dimension. Finally, a probabilistic neural network (PNN) coupled with entropic thresholding techniques is developed for mass extraction. Since the shapes of masses are crucial in classification between benignancy and malignancy, four shape features are further generated and joined with the five features previously used in mass detection to be implemented in another PNN for mass classification. To evaluate our designed system a data set collected in the Taichung Veteran General Hospital, Taiwan, R.O.C. was used for performance evaluation. The results are encouraging and have shown promise of our system.

关键词： computer aided analysis

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：