检索结果-内蒙古大学图书馆

1st international conference on Advanced Technologies for Signal and image processing (ATSIP)

作者： Faleh, Rabeb Othman, Mehdi Kachouri, Abdennaceur Aguir, Khalifa Univ Sfax Natl Sch Engineer Sfax LETI Lab Dept Elect Engn Sfax Tunisia Aix Marseille Univ CNRS IM2NP UMR 7334 F-13397 Marseille France Univ Gabes Higher Inst Ind Syst Gabes ISSIG Gabes 6011 Tunisia

ISBN: (纸本)9781479948888

In this current paper, a sensor array based on WO3 gas sensors has been used to detect O-3. Comparing to the classical analysis methods, metal oxide sensors present the advantages of low cost, short response time and versatility. However, the use of these sensors is limited due to their lack of selectivity. Representative variable as gas response and time response have been extracted from gas response base. Considering concentration range from 50 ppb to 400 ppb, database is established for four gas sensor. Then, the variables are grouped in database and tested with principal component analysis (PCA) to evaluate the contribution to the classification to enhance the sensors selectivity.

关键词： e-nose gas sensor selectivity Tungsten oxide pattern analysis principal component analysis(PCA)

来源：评论

学校读者我要写书评

暂无评论

An Improved Separating Hyperplane Method with Application to Embedded Intelligent Devices 21

引用

21st international conference on Neural Information processing (ICONIP)

作者： Li, Yanjun Guo, Ping Xin, Xin Beijing Inst Technol Sch Comp Sci & Technol Beijing 100081 Peoples R China Beijing Normal Univ Image Proc & Pattern Recognit Lab Beijing 100875 Peoples R China

ISBN: (纸本)9783319126401

Classification is a common task in pattern recognition. Classifiers used in embedded intelligent devices need a good trade-off between prediction accuracy, resource consumption and prediction speed. Support vector machine(SVM) is accurate but its run-time complexity is higher due to the large number of support vectors. A new separating hyperplane method (NSHM) for the binary classification task was proposed. NSHM allows fast classification. However, NSHM is order-sensitive and this affects its classification accuracy. Inspired by NSHM, we propose CSHM, a combining separating hyperplane method. CSHM combines all optimal separating hyperplanes found by NSHM. Experimental results on UCI machine learning Repository show that, compared with NSHM and SVM, CSHM achieves a better trade-off between prediction accuracy, resource consumption and prediction speed.

关键词： separating hyperplane embedded intelligent devices support vector machines

来源：评论

学校读者我要写书评

暂无评论

Dictionary learning-Based Volumetric image Classification for the Diagnosis of Age-Related Macular Degeneration

Dictionary Learning-Based Volumetric Image Classification fo...

引用

10th international conference on machine learning and Data Mining (MLDM)

作者： Albarrak, Abdulrahman Coenen, Frans Zheng, Yalin Univ Liverpool Dept Comp Sci Liverpool L69 3BX Merseyside England Univ Liverpool Dept Eye & Vis Sci Liverpool L69 3BX Merseyside England

ISBN: (纸本)9783319089799;9783319089782

A discriminative dictionary-based approach to supporting the classification of 3D Optical Coherence Tomography (OCT) retinal images, so as to determine the presence of Age-related Macular Degeneration (AMD), is described. AMD is one of the leading causes of blindness in people aged over 50 years. The proposed approach is founded on the concept of a uniform 3D image decomposition into a set of sub-volumes where each sub-volume is described in terms of a "spatial gradient" histogram, which in turn is used to define a set of feature vectors (one per sub-volume). Feature selection is conducted using the maximum sum of the squared values of each feature vector for each sub-volume. After that, a "coding-pooling" framework is applied so that each image is represented as a single feature vector. The "coding-pooling" framework generates a representative subset of feature vectors called a dictionary, and then use this dictionary as a guide for the generation of a single feature vectors for each volume. Experiments conducted using the proposed approach, in comparison with range of alternatives, indicated that the approach outperformed other existing methods with an accuracy of 95.2%, sensitivity of 95.7% and specificity of 94.6%.

关键词： Data mining image decomposition Spatial gradient histograms diactioanry learning Medical image processing Optical Coherence Tomography

来源：评论

学校读者我要写书评

暂无评论

Generative embeddings based on Rician mixtures for kernel-based classification of magnetic resonance images

引用

NEUROCOMPUTING 2014年 123卷 49-59页

作者： Carli, Anna C. Figueiredo, Mario A. T. Bicego, Manuele Murino, Vittorio Univ Verona Dipartimento Informat I-37100 Verona Italy Inst Super Tecn Inst Telecomunicacoes Lisbon Portugal IIT PAVIS Pattern Anal & Comp Vis Dept Genoa Italy

Classical approaches to classifier learning for structured objects (such as images or sequences) are based on probabilistic generative models. On the other hand, state-of-the-art classifiers for vectorial data are learned discriminatively. In recent years, these two dual paradigms have been combined via the use of generative embeddings (of which the Fisher kernel is arguably the best known example);these embeddings are mappings from the object space into a fixed dimensional score space, induced by a generative model learned from data, on which a (maybe kernel-based) discriminative approach can then be used. This paper proposes a new semi-parametric approach to build generative embeddings for classification of magnetic resonance images (MRI). Based on the fact that MRI data is well described by Rice distributions, we propose to use Rician mixtures as the underlying generative model, based on which several different generative embeddings are built. These embeddings yield vectorial representations on which kernel-based support vector machines (SVM) can be trained for classification. Concerning the choice of kernel, we adopt the recently proposed nonextensive information theoretic kernels. The methodology proposed was tested on a challenging classification task, which consists in classifying MRI images as belonging to schizophrenic or non-schizophrenic human subjects. The classification is based on a set of regions of interest (ROIs) in each image, with the classifiers corresponding to each ROI being combined via AdaBoost. The experimental results show that the proposed methodology outperforms the previous state-of-the-art methods on the same dataset. (C) 2013 Elsevier B.V. All rights reserved.

关键词： Rician mixture EM algorithm Generative embedding Discriminative learning Information theory Boosting

来源：评论

学校读者我要写书评

暂无评论

Detections of Intima-Media Thickness in B-Mode Carotid Artery images Using Segmentation Methods 1

引用

1st international conference on Internet Computing and Information Communications (ICICIC)

作者： Savithri, V. Purushothaman, S. Mother Teresa Womens Univ Dept Comp Sci Kodaikanal India PET Engn Coll Tirunelveli Tamil Nadu India

ISBN: (纸本)9788132212980;9788132212997

This study presents the investigations carried out on carotid artery to identify the intima-media thickness of carotid artery that affected with plaques. B-mode ultrasound image video of the artery has been used as the data for processing. The frames of the video are processed to know the plaque properties of the artery. In order to achieve this, two segmentation processing techniques have been used on each frame. The features extracted from the frames are consolidated to know the conditions of the artery. Information of a frame are converted into features. The values of the features are estimated by artificial neural network (ANN) algorithm. ANN has not been used extensively by the past. ANN is used in estimating the plaque thickness in the carotid artery.

关键词： B-mode Back propagation network pattern recognition Artificial neural network Clustering

来源：评论

学校读者我要写书评

暂无评论

Object Segmentation through Multiple Instance learning

Object Segmentation through Multiple Instance Learning

引用

6th international conference on image and Signal processing (ICISP)

作者： Gondra, Iker Xu, Tao Chiu, David K. Y. Cormier, Michael St Francis Xavier Univ Dept Comp Sci Antigonish NS B2G 1C0 Canada Univ Guelph Sch Comp Sci Guelph ON Canada Univ Waterloo Sch Comp Sci Waterloo ON Canada

ISBN: (纸本)9783319079981;9783319079974

An object of interest (OOI) in an image usually consists of visually coherent regions that, together, encompass the entire OOI. We use Multiple Instance learning (MIL) to determine which regions in an over-segmented image are part of the OOI. In the learning stage, a set of over-segmented images containing, i.e., positive, and not containing, i.e., negative, an instance of the OOI is used as training data. The resulting learned prototypes represent the visual appearances of OOI regions. In the OOI segmentation stage, the new image is over-segmented and regions that match prototypes are merged. Our MIL method does not require prior knowledge about the number of regions in the OOI. We show that, with the coexistence of multiple prototypes corresponding to the regions of the OOI, the maxima of the formulation are good estimates of such regions. We present initial results over a set of images with a controlled, relatively simple OOI.

关键词： image segmentation object recognition multiple instance learning diverse density adaptive kernel

来源：评论

学校读者我要写书评

暂无评论

Real-time Eye Tracking and Event Identification Techniques for Smart TV applications 1

Real-time Eye Tracking and Event Identification Techniques f...

引用

1st IEEE international conference on Consumer Electronics - Taiwan (ICCE-TW)

作者： Chen, Yen-Lin Chiang, Chuan-Yen Yu, Chao-Wei Sun, Wei-Chen Yuan, Shyan-Ming Natl Taipei Univ Technol Dept Comp Sci & Informat Engn Taipei 106 Taiwan Natl Chiao Tung Univ Inst Comp Sci & Engn Hsinchu Taiwan

ISBN: (纸本)9781479948512

Due to the progress of computer technology, the detecting accuracy and speed of pattern recognition are promoted. Additionally, according to research advances in embedded systems, the applications of digital cameras, such as internet cameras, smart phones and smart TV, are widely used in nowadays. Therefore, we propose a set of real-time human-eye recognition and tracking systems with human-computer interaction mechanism. It can be adopted to help the smart TV users or any device which have camera and computing power to improve their using experience. The proposed system is demonstrated on OMAP4430 for embedded applications, and experimental results show that the proposed architecture is capable of effective and real-time eye position tracking. The proposed system can also detect the status of eye blink. Even for the case of uneven lighting, the proposed system can successfully recognize the human eyes with high accuracy.

关键词： Computer architecture Computer vision Human computer interaction image sequences Real-time systems Sensors TV Television image sequences computer human interaction Computer Architecture real-time systems images computer vision Computer technology Sensors Blink Reflexes TRACKING SYstEM embedded application Cameras Digital cameras

来源：评论

学校读者我要写书评

暂无评论

An Low-Power Microcontroller with Accuracy-Controlled Signal-to-Event Converter for Rare-Event Human Activity-Sensing Applications 1

An Low-Power Microcontroller with Accuracy-Controlled Signal...

引用

1st IEEE international conference on Consumer Electronics - Taiwan (ICCE-TW)

作者： Park, Daejin Kim, Tag Gon Gumi Elect & Informat Technol Res Inst Biomed IT Convergence Ctr Gumi South Korea Korea Adv Inst Sci & Technol Dept Elect Engn Taejon South Korea

ISBN: (纸本)9781479948512

A specially designed event-driven sensor processor architecture is proposed to implement the low-power microcontroller for rare-event activity sensing applications. The rare-event applications have a property of very long event-to-event distance like human-triggered events, for which the accuracy error of the signal detection is allowed. The early evaluation for the fingerprint of the incoming sensor signal is performed by the proposed signal-to-event converter, which enables the entire post-processing unit only handles the atomic events. The proposed architecture is implemented with additional 7500 NAND gates and 1 KB SRAM tracer in 0.18um CMOS process by adding the proposed building blocks on the commercial 8051-based microcontroller to reduce the sensor data-processing current, consuming only 20% compared the conventional discrete-time signal analysis.

关键词： Computer architecture Logic gates Microcontrollers pattern matching Random access memory Sensors Signal analysis Microcontrollers Signal analysis random access machine pattern matching Computer Architecture Logic gates Sensors LOW POWER NAND GATES signal acquisition tellurium Converters

来源：评论

学校读者我要写书评

暂无评论

Automated page layout simplification of patrologia graeca 14

Automated page layout simplification of patrologia graeca

引用

1st international conference on Digital Access to Textual Cultural Heritage, DATeCH 2014

作者： Robertson, Bruce Dalitz, Christoph Schmitt, Fabian Department of Classics Mount Allison University 63D York St. #414 Hart Hall Sackville NB Canada Institute for Pattern Recognition Niederrhein University of Applied Sciences Reinarzstr. 49 47805 Krefeld Germany

ISBN: (纸本)9781450325882

This paper illustrates how automated image preprocessing can improve OCR results of an important 19th century series of Greek authors. The roughly 75,000 ancient Greek pages from the volumes published by J.-P. Migne known today as Patrologia Graeca have a wholly regular layout, but one that is not easily reduced to a reading order by standard layout-analysis algorithms. We use a Hough transform and k-nearest-neighbor distance rejection to locate and then remove the features that confuse modern layout analysis. This allows us to separate the valuable Greek text from the less useful Latin translation, greatly simplifying later stages of processing, including de-hyphenation, automatic spell-checking and editing. Our implementation is based upon the Gamera Python package. We make the source code of our solution freely available.

关键词： Hough transforms

来源：评论

学校读者我要写书评

暂无评论

A novel edge detection and pattern recognition algorithm based on beamlet transform theory for a vision-based Wheeled Mobile Robot

A novel edge detection and pattern recognition algorithm bas...

引用

1st international conference on Emerging Trends in Powering the Nation, ICETPN 2014

作者： Soumya, S. Geethapriya, V. Department of EEE Rajalakshmi Engineering College Thandalam Chennai India

ISBN: (纸本)9783038350965

Number of strategies have been deployed to make Wheeled Mobile Robots (WMRs) independent or with limited dependence on human assistance/intervention. In order to enhance their mobility, improvements have been made in their pattern recognition and path planning strategies. Vision-based approach is preferred over stereopsis-based, laser-range finders and GPS-based approaches on account of higher resolution and longer ranges. Vision-Based Wheeled Mobile Robot(VB-WMR) uses an active vision system where the observer is in motion. The goal of this project is to provide a novel algorithm aimed to detect and recognize direction indicators on sign boards placed on the course of the WMR. Here visual processing is devoted to noting the salient components like the edges of patterns on sign boards rather than processing the entire image. The proposed algorithm is based on the beamlet theory, which retains directional information besides the location and scale and performs well under noisy conditions. The developed algorithm in MATLAB is later to be incorporated with the VB-WMR. © (2014) Trans Tech Publications, Switzerland.

关键词： Edge detection

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：