检索结果-内蒙古大学图书馆

IEEE Symposium on Computational Intelligence and Data Mining (CIDM)

作者： Yangchi Chen Melba M. Crawford Joydeep Ghosh Center for Space Research University of Texas Austin Austin TX USA Laboratory for Applications of Remote Sensing Lilly Hall of Life Sciences Purdue University USA Department of Electrical and Computer Engineering University of Texas Austin USA

Hyperspectral data provide new capability for discriminating spectrally similar classes, but unfortunately such class signatures often overlap in multiple narrow bands. Thus, it is useful to incorporate reliable spatial information when possible. However, this can result in increased dimensionality of the feature vector, which is already large for hyperspectral data. Markov random field (MRF) approaches, such as iterated conditional modes (ICM), can provide evidence relative to the class of a neighbor through Gibbs' distribution, but suffer from computational requirements and curse of dimensionality issues when applied to hyperspectral data. In this paper, a new knowledge based stacking approach is presented to utilize spatial information within homogeneous regions and at class boundaries, while avoiding the curse of dimensionality. The approach learns the location of the class boundary and combines original bands with the extracted spectral information of a neighborhood to train a hierarchical support vector machine (HSVM) classifier. The new method is applied to hyperspectral data collected by the Hyperion sensor on the EO-1 satellite over the Okavango delta of Botswana. Classification accuracies are compared to those obtained by a pixel-wise HSVM classifier, majority filtering and ICM to demonstrate the advantage of the knowledge based stacking approach.

关键词： Stacking Hyperspectral imaging Hyperspectral sensors Narrowband Markov random fields Distributed computing Data mining Support vector machines Support vector machine classification Satellites

来源：评论

学校读者我要写书评

暂无评论

Multiresolution manifold learning for classification of hyperspectral data

Multiresolution manifold learning for classification of hype...

引用

IEEE International Symposium on Geoscience and remote sensing (IGARSS)

作者： Wonkook Kim Yangchi Chen Melba M. Crawford James C. Tilton Joydeep Ghosh Laboratory for Applications of Remote Sensing Purdue University USA Center for Space Research University of Texas Austin USA National Aeronautics and Space Administration Information Science and Technology Research Department of Electrical and Computer Engineering University of Texas Austin USA

Nonlinear manifold learning algorithms assume that the original high dimensional data actually lie on a low dimensional manifold defined by local geometric distances between samples. Most of the traditional methods have focused only on the spectral distances in calculating the local dissimilarity of samples, whereas in the case of image data, the spatial distribution and localized contextual information of image samples could provide useful information. As a framework for integrating spatial and spectral information associated with image samples, a hierarchical spatial-spectral segmentation method is investigated for constructing the manifold structure. The new approach, which develops the manifold for the purpose of classification, incorporates an updating scheme whereby the spatial information and class labels are transferred through the segmentation hierarchy. It is applied to hyperspectral data collected by the Hyperion sensor on the EO-1 satellite over the Okavango Delta of Botswana. Classification accuracies and generalization capability are compared to those achieved by the best basis binary hierarchical classifier, the hierarchical support vector machine classifier, and the shortest path k-nearest neighbor classifier.

关键词： Hyperspectral imaging image segmentation Hyperspectral sensors Space technology Manifolds Spatial resolution Application software remote sensing Information science Satellites

来源：评论

学校读者我要写书评

暂无评论

Comparison of modeled and observed microwave emissivities of water surfaces in the presence of breaking waves and foam

Comparison of modeled and observed microwave emissivities of...

引用

IEEE International Symposium on Geoscience and remote sensing (IGARSS)

作者： S. Padmanabhan S. C. Reising W. E. Asher V. Raizer P. W. Gaiser Department of Electrical and Computer Engineering Colorado State University Fort Collins CO USA Applied Physics Laboratory University of Washington Seattle WA USA Zel Technologies Limited Liability Company Fairfax VA USA Remote Sensing Division Naval Research Laboratory Inc. WA USA

Wind speed has been retrieved reliably from SSM/I microwave radiometric measurements since 1990 with a precision of better than 2 m/s. However, operational requirements of the National Polar-orbiting Operational Environmental Satellite System (NPOESS) dictate retrieval of the ocean surface wind vector, i.e. both speed and direction. This requires improved understanding and forward modeling of the physical processes governing ocean surface microwave emissivity. Although the microwave brightness temperature over the ocean exhibits a relatively strong dependence on wind speed, the dependence on wind direction with respect to the azimuthal angle of observation is at most a few Kelvin peak-to-peak. Although it is known that breaking waves and foam significantly affect the microwave emission of the ocean surface at 6.8, 10.8, 19 and 37 GHz, the effect of foam on the azimuthal dependence of sea surface emissivity is not well understood. This understanding is critical to improve forward models to achieve the accuracy required for ocean surface wind vector retrievals. Recently, electromagnetic modeling of the emissivity of foam generated by breaking waves has been improved by considering the vertical and horizontal heterogeneity of thick layers of foam at the ocean- atmosphere interface [1]. In this work, we compare foam emissivities both calculated using this model with those inferred from microwave brightness temperature measurements performed during the Polarimetric Observations of the Emissivity of Whitecaps Experiment (POEWEX'04).

关键词： Surface waves Sea surface Ocean temperature Wind speed Microwave measurements Sea measurements Brightness temperature Atmospheric modeling Microwave radiometry Velocity measurement

来源：评论

学校读者我要写书评

暂无评论

Characterization of anomaly detection in hyperspectral imagery

引用

Sensor Review 2006年第2期26卷 137-146页

作者： Chang, Chein-I Hsueh, Mingkai Remote Sensing Signal and Image Processing Laboratory Department of Computer Science and Electrical Engineering University of Maryland Baltimore MD United States

Purpose - The paper aims to characterize anomaly detection in hyperspectral imagery. Design/methodology/approach - This paper develops an adaptive causal anomaly detector (ACAD) to investigate several issues encountered in hyperspectral image analysis which have not been addressed in the past. It also designs extensive synthetic image-based computer simulations and real image experiments to substantiate the work proposed in this paper. Findings - This paper developed an ACAD and custom-designed computer simulations and real image experiments to successfully address several issues in characterizing anomalies for detection, which are - first, how large size for a target to be considered as an anomaly? Second, how an anomaly responds to its proximity? Third, how sensitive for an anomaly to noise? Finally, how different anomalies to be detected? Additionally, it also demonstrated that the proposed ACAD can be implemented in real time processing and implementation. Originality/value - This paper is the first work on investigation of several issues related to anomaly detection in hyperspectral imagery via extensive synthetic image-based computer simulations and real image experiments. In addition, it also develops a new developed an ACAD to address these issues and substantiate its performance.

关键词： image processing

来源：评论

学校读者我要写书评

暂无评论

NONPARAMETRIC ENTROPY-BASED COUPLED MULTI-SHAPE MEDICAL image SEGMENTATION

NONPARAMETRIC ENTROPY-BASED COUPLED MULTI-SHAPE MEDICAL IMAG...

引用

IEEE International Symposium on Biomedical Imaging

作者： Alireza Akhoundi-Asl Hamid Soltanian-Zadeh School of Cognitive Sciences Institute of Studies in Theoretical Physics and Mathematics Tehran Iran Control and Intelligent Processing Center of Excellence (CIPCE) School of Electrical and Computer Engineering University of Tehran Tehran Iran Image Analysis Laboratory Radiology Department Henry Ford Health System Detroit MI USA

ISBN: (纸本)9781424406715;1424406714

We propose a 3D nonparametric, entropy-based, coupled, multishape approach for the segmentation of subcortical brain structures in magnetic resonance images (MRI). Our method uses PCA to capture structures variability. Because of complex relationships of pose and shape of the coupled structures, we only use their shape and size relation. To this end, we apply separate registrations of the structures. For each structure, we consider a similarity transform using seven parameters. In addition, to generate most accurate results, we estimate probability density functions (pdf) iteratively. The proposed method minimizes an entropy-based energy function using quasi-Newton algorithm. To improve the results, we use analytical derivatives. Sample results are given for the segmentation of putamen, thalamus and caudate illustrating the impact of coupling on the accuracy of the results.

关键词： Biomedical imaging image segmentation Shape Couplings Brain Magnetic resonance Magnetic resonance imaging Principal component analysis Probability density function Iterative algorithms

来源：评论

学校读者我要写书评

暂无评论

Using context and a genetic algorithm for knowledge-assisted image analysis

Using context and a genetic algorithm for knowledge-assisted...

引用

1st International Conference on Semantic and Digital Media Technologies, SAMT 2006

作者： Dasiopoulou, Stamatia Papadopoulos, George Th. Mylonas, Phivos Avrtihis, Yannis Kompatsiaris, Ioannis Information Processing Laboratory Electrical and Computer Engineering Department Aristotle University Thessaloniki Greece Informatics and Telematics Institute Centre for Research and Technology Greece Image Video and Multimedia Laboratory School of Electrical and Computer Engineering National Technical University Athens Greece

In this poster, we present an approach to contex-tualized semantic image annotation as an optimization problem. Ontologies are used to capture general and contextual knowledge of the domain considered, and a genetic algorithm is applied to realize the final annotation. Experiments with images from the beach vacation domain demonstrate the performance of the proposed approach and illustrate the added value of utilizing contextual information.

关键词： Ontology

来源：评论

学校读者我要写书评

暂无评论

Super-resolution in the presence of space-variant blur

Super-resolution in the presence of space-variant blur

引用

International Conference on Pattern Recognition

作者： K.V. Suresh A.N. Rajagopalan Image Processing and Computer Vision Laboratory Department of Electrical Engineering IIT Madras Chennai India

An efficient algorithm using maximum a posteriori-Markov random field (MAP-MRF) based approach for recovering a high-resolution image from multiple sub-pixel shifted low-resolution images is proposed. The algorithm can be used for super-resolution of both space-invariant and space-variant blurred images. We prove an important theorem that the posterior is also Markov and derive the exact posterior neighborhood structure in the presence of warping, blurring and down-sampling operations. The posterior being Markov enables us to perform all matrix operations as local image domain operations thereby resulting in a considerable speedup. Experimental results are given to demonstrate the effectiveness of our method

关键词： Spatial resolution image resolution Cameras image processing computer vision Laboratories Layout Frequency domain analysis Markov random fields Reconstruction algorithms

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Unmixing of Hyperspectral imagery

Unsupervised Unmixing of Hyperspectral Imagery

引用

Midwest Symposium on Circuits and Systems (MWSCAS)

作者： Yahya M. Masalmah Miguel Velez-Reyes Laboratory for Applied Remote Sensing and Image ProcessingElectrical and Computer Engineering Department University of Puerto Rico Mayaguez Puerto Rico

This paper presents an approach for simultaneous determination of end members and their abundances in hyperspectral imagery using a constrained positive matrix factorization. The algorithm presented here solves the constrained PMF using Gauss-Seidel method. This algorithm alternates between the end members matrix updating step and the abundance estimation step until convergence is achieved. Preliminary results using a subset of the Enrique Reef image data are presented. These results show the potential of the method to solve the unsupervised unmixing problem.

关键词： Hyperspectral imaging Hyperspectral sensors Spatial resolution Pixel remote sensing Spectroscopy Laboratories image processing Gaussian processes Convergence

来源：评论

学校读者我要写书评

暂无评论

Web Document image Retrieval System Based on Word Spotting

Web Document Image Retrieval System Based on Word Spotting

引用

IEEE International Conference on image processing

作者： K. Zagoris N. Papamarkos C. Chamzas Image Processing and Multimedia Laboratory Department of Electrical & Computer Engineering Democritus University of Thrace Xanthi Greece

Nowadays, the huge non-indexing quantities of image archives (especially document images) require the development of intelligent tools for their retrieval with convenience comparable of the texts search engines. The proposed technique addresses the document retrieval problem by a word matching procedure. It performs matching directly in the images bypassing OCR and using word-images as queries. It is constituted of two different parts: The offline and the online operation. In the offline operation, the archive of document images is examined and the results are stored in a database. The online operation consists of the Web interface, the creation of the word's image and finally, the matching stage. The proposed matching process it can be described shortly as a two-threshold rating system. Finally, the proposed system has been build and it can be found in at the Web address: http://***/irs2.

关键词： image retrieval image databases Search engines Optical character recognition software Information retrieval Feature extraction Spatial databases image processing Multimedia systems Laboratories

来源：评论

学校读者我要写书评

暂无评论

Improving Shape from Focus Using Defocus Information

Improving Shape from Focus Using Defocus Information

引用

International Conference on Pattern Recognition

作者： K.S. Pradeep A.N. Rajagopalan Image Processing and Computer Vision Laboratory Department of Electrical Engineering Indian Institute of Technology Madras India

ISBN: (纸本)0769525210

Shape from focus (SFF) method determines the degree of focus in a sequence of observations to estimate the shape of a 3-D object. Existing SFF algorithms use an ad hoc interpolation strategy to account for the error due to the finite step-size by which the translational table is moved while capturing the images. We propose an improved SFF method that uses relative defocus blur derived from actual image data to arrive at the final estimates of the shape of the object. A space-variant image restoration scheme is also proposed to obtain a focused image of the 3-D object. The shape estimates as well as the quality of the restored image using the proposed method are superior to that of traditional SFF

关键词： Shape Focusing Cameras Lenses image restoration Interpolation Pixel Polynomials image sensors image processing

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：