Filters








127,762 Hits in 6.4 sec

Unsupervised Spectral-Spatial Feature Selection-Based Camouflaged Object Detection Using VNIR Hyperspectral Camera

Sungho Kim
2015 The Scientific World Journal  
This letter proposes a fully autonomous feature selection and camouflaged object detection method based on the online analysis of spectral and spatial features.  ...  Conventional supervised learning methods for hyperspectral images can be a feasible solution. Such approaches, however, require a priori information of a camouflaged object and background.  ...  A camouflaged object detection problem can be regarded as selecting suitable spectral bands that discriminate interesting region in normal background.  ... 
doi:10.1155/2015/834635 pmid:25879073 pmcid:PMC4386716 fatcat:dn6muvgp4rhh7e2csq4jwinsrq

A Hierarchical Object Recognition System Based on Multi-scale Principal Curvature Regions

Wei Zhang, Hongli Deng, T.G. Dietterich, E.N. Mortensen
2006 18th International Conference on Pattern Recognition (ICPR'06)  
Each region is described by an intensity-based statistical descriptor and a PCA-SIFT descriptor. The spatial relations between regions are represented by a cluster-index distribution histogram.  ...  most discriminative features in each layer.  ...  We would like to thank Caltech Vision Lab and Yan Ke for providing test dataset and source code for descriptor.  ... 
doi:10.1109/icpr.2006.1195 dblp:conf/icpr/ZhangDDM06 fatcat:dugiuqywhraarfbqed7cmaqovm

Computationally Intelligent Methods for Mining 3D Medical Images [chapter]

Despina Kontos, Vasileios Megalooikonomou, Fillia Makedon
2004 Lecture Notes in Computer Science  
We focus on detecting discriminative Regions of Interest (ROIs) and mining associations between their spatial distribution and other clinical assessment.  ...  To identify these highly informative regions, we propose utilizing statistical tests to selectively partition the 3D space into a number of hyper-rectangles.  ...  Acknowledgement The authors would like to thank A. Saykin for providing the fMRI data set and clinical expertise and J. Ford for performing some of the preprocessing of this data set.  ... 
doi:10.1007/978-3-540-24674-9_9 fatcat:f44lzhgo2bcflay4fg7rtc376u

An MCMC Feature Selection Technique for Characterizing and Classifying Spatial Region Data [chapter]

Despina Kontos, Vasileios Megalooikonomou, Marc J. Sobel, Qiang Wang
2004 Lecture Notes in Computer Science  
We propose a novel statistical approach based on a supervised framework for reducing the dimensionality of the initial feature space, selecting the most discriminative features.  ...  We focus on characterizing spatial region data when distinct classes of structural patterns are present.  ...  Acknowledgement The authors would like to thank A. Saykin for providing the fMRI dataset and clinical expertise and H. Dutta for working on preliminary experiments.  ... 
doi:10.1007/978-3-540-27868-9_40 fatcat:6xtz2pg47nfrxgltzflkd6icmy

An adaptive partitioning approach for mining discriminant regions in 3D image data

Vasileios Megalooikonomou, Despina Kontos, Dragoljub Pokrajac, Aleksandar Lazarevic, Zoran Obradovic
2007 Journal of Intelligent Information Systems  
In this paper, we propose a novel approach for detecting spatial regions that are highly discriminative among different classes of three dimensional (3D) image data.  ...  The main idea of our approach is to treat the initial 3D image as a hyper-rectangle and search for discriminative regions by adaptively partitioning the space into progressively smaller hyperrectangles  ...  Acknowledgements The authors would like to thank A. Saykin for providing the fMRI dataset and clinical expertise.  ... 
doi:10.1007/s10844-007-0043-2 fatcat:w4t4zt5n75b3vjnvjf7urbcjui

On spatial selectivity and prediction across conditions with fMRI [article]

Yannick Schwartz, Gaël Varoquaux, Bertrand Thirion (INRIA Saclay - Ile de France, LNAO)
2012 arXiv   pre-print
Instead, we propose to use the full statistical images to define regions of interest (ROIs).  ...  We provide some preliminary quantification of these similarities, and show that selection transfer makes it possible to set a spatial scale yielding ROIs that are more specific to the context of interest  ...  Experimental results for transfer learning We are interested in transfer learning: we learn a discriminative model on the source task with a univariate feature selection, and predict the labels on the  ... 
arXiv:1209.1450v1 fatcat:5rjohxc5urflbiadtb2dm3kyjq

On Spatial Selectivity and Prediction across Conditions with fMRI

Yannick Schwartz, Gael Varoquaux, Bertrand Thirion
2012 2012 Second International Workshop on Pattern Recognition in NeuroImaging  
Instead, we propose to use the full statistical images to define regions of interest (ROIs).  ...  We provide some preliminary quantification of these similarities, and show that selection transfer makes it possible to set a spatial scale yielding ROIs that are more specific to the context of interest  ...  Experimental results for transfer learning We are interested in transfer learning: we learn a discriminative model on the source task with a univariate feature selection, and predict the labels on the  ... 
doi:10.1109/prni.2012.24 dblp:conf/prni/SchwartzVT12 fatcat:uyae26qbgfgprflqt645j7cowe

Classification of Structural Images via High-Dimensional Image Warping, Robust Feature Extraction, and SVM [chapter]

Yong Fan, Dinggang Shen, Christos Davatzikos
2005 Lecture Notes in Computer Science  
with respect to the number of selected features and the SVM kernel size.  ...  A morphological representation of the anatomy of interest is first obtained using highdimensional template warping, from which regions that display strong correlations between morphological measurements  ...  This requires a feature selection method to select a small set of the most informative features for classification.  ... 
doi:10.1007/11566465_1 fatcat:jbhpuojf7nd5pnoa7s3xz72cti

Extracting Discriminative Information from Medical Images: A Multivariate Linear Approach

Carlos Thomaz, Nelson Aguiar, Sergio Oliveira, Fabio Duran, Geraldo Busatto, Duncan Gillies, Daniel Rueckert
2006 Computer Graphics and Image Processing (SIBGRAPI), Proceedings of the Brazilian Symposium on  
Statistical discrimination methods are suitable not only for classification but also for characterisation of differences between a reference group of patterns and the population under investigation.  ...  The conceptual and mathematical simplicity of the approach, which pivotal step is spatial normalisation, involves the same operations irrespective of the complexity of the experiment or nature of the data  ...  The idea of using PCA plus an LDA-based approach to discriminate patterns of interest is not new.  ... 
doi:10.1109/sibgrapi.2006.19 dblp:conf/sibgrapi/ThomazAODBGR06 fatcat:cfmuqhy3ojahtkjxgq2tnyb6uq

Texture Driven Hierarchical Fusion for Multi-Biometric Sys-tem

Devendra Reddy Rachapalli, Hemantha Kumar Kalluri
2018 International Journal of Engineering & Technology  
Finally, t-normalized feature level fusion is incorporated as a last stage to create the most reliable template for the identification process.  ...  Texture feature metrics are extracted from multi-level texture regions and principal component analyzes are evaluated to select potentially useful texture values during template creation.  ...  For ternary pattern upper limit the interested regions are 1 as pixel values and for ternary pattern lower limit the interested regions are -1 as pixel values.  ... 
doi:10.14419/ijet.v7i4.24.21766 fatcat:ewj6julfzzamffl2xeozq4252u

High Classification Accuracy for Schizophrenia with Rest and Task fMRI Data

Wei Du, Vince D. Calhoun, Hualiang Li, Sai Ma, Tom Eichele, Kent A. Kiehl, Godfrey D. Pearlson, Tülay Adali
2012 Frontiers in Human Neuroscience  
Several components, including DMN, temporal, and medial visual regions, are consistently present in the set of features that yield high classification accuracy.  ...  By combining a two-level feature identification scheme with kernel principal component analysis (KPCA) and Fisher's linear discriminant analysis (FLD), we achieve high classification rates in discriminating  ...  We select 14 components of interest as features to discriminate healthy controls and patients in each data set.  ... 
doi:10.3389/fnhum.2012.00145 pmid:22675292 pmcid:PMC3366580 fatcat:7q55jylunzdftge4d6cq4bujme

Texture Analysis in Magnetic Resonance Imaging: Review and Considerations for Future Applications [chapter]

Andrés Larroza, Vicente Bodí, David Moratal
2016 Assessment of Cellular and Organ Function and Dysfunction using Direct and Derived MRI Methodologies  
In the present chapter, we describe texture analysis as a process consisting of six steps: MRI acquisition, region of interest (ROI) definition, ROI preprocessing, feature extraction, feature selection  ...  Texture analysis is a technique used for the quantification of image texture.  ...  Figure 5 shows examples of the three aforementioned ROI definition approaches. Figure 5 . Approaches for defining a region of interest (ROI) over a brain tumor.  ... 
doi:10.5772/64641 fatcat:sw7nodso7rfpfjohaopmrt2zta

Joint Spatial Denoising and Active Region of Interest Delineation in Functional Magnetic Resonance Imaging

Bernard Ng, Rafeef Abugharbieh, Samantha J. Palmer, Martin J. McKeown
2007 IEEE Engineering in Medicine and Biology Society. Conference Proceedings  
The validity of the proposed method is suggested by the fact that using one feature for denoising (e.g. spatial variance) results in greater effect size in another feature (e.g. average activation statistics  ...  In region of interest (ROI) based functional magnetic resonance imaging (fMRI) group analysis, errors in delineation of an ROI or inclusion of non-active voxels within an ROI can bias the statistical results  ...  An alternative approach is to manually draw regions of interest (ROIs) individually for each subject, and examine the statistical properties of regional activation across subjects.  ... 
doi:10.1109/iembs.2007.4353062 pmid:18002728 fatcat:didqt5ktsrgkdlnfegrbcl7qpy

AUTOMATIC CLASSIFICATION OF STRUCTURAL MRI FOR DIAGNOSIS OF NEURODEGENERATIVE DISEASES

Gloria Diaz, Eduardo Romero, Juan Antonio Hernández-Tamames, Vicente Molina, Norberto Malpica
2010 Acta Biológica Colombiana  
The method uses the deformation values from a set of regions, automatically identified as relevant, in a process that selects the statistically significant regions of a t-test under the restriction that  ...  this significance must be spatially coherent within a neighborhood of 5 voxels.  ...  A SVM-RFE approach is used for selecting the most discriminative features that are finally used for training a SVM learning model, which classifies individual subjects.  ... 
doaj:5b986420968741458b82032e1ca95fea fatcat:u7vs7emqarfi3cw67pq5xjtbde

Regions of interest for accurate object detection

P. Kapsalas, K. Rapantzikos, A. Sofou, Y. Avrithis
2008 2008 International Workshop on Content-Based Multimedia Indexing  
The basic idea of the approach is that object location can be determined by clustering points of interest and hierarchically forming candidate regions according to similarity and spatial proximity predicates  ...  In this paper we propose an object detection approach that extracts a limited number of candidate local regions to guide the detection process.  ...  The training process uses AdaBoost to select a subset of features and construct the classifier. The classifier consists of a linear combination of the selected features.  ... 
doi:10.1109/cbmi.2008.4564940 dblp:conf/cbmi/KapsalasRSA08 fatcat:fmrcv72x4jdezipsxhobxu5o3m
« Previous Showing results 1 — 15 out of 127,762 results