Filters








309 Hits in 3.7 sec

Learning Markov Blankets for Continuous or Discrete Networks via Feature Selection [chapter]

Houtao Deng, Saylisse Davila, George Runger, Eugene Tuv
2011 Studies in Computational Intelligence  
We compare our algorithm in the causal structure learning problem to other well-known feature selection methods, and to a Bayesian local structure learning algorithm.  ...  Then an ensemble feature selection method is used to learn Markov blankets for either discrete or continuous networks (without linear, Gaussian assumptions).  ...  that generated the data, to facilitate data visualization and data understanding [6] .  ... 
doi:10.1007/978-3-642-22910-7_7 fatcat:joemcoidjzgwzi2vaujvva2qhq

Unsupervised Band Selection for Hyperspectral Imagery Classification Without Manual Band Removal

Sen Jia, Zhen Ji, Yuntao Qian, Linlin Shen
2012 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Then affinity propagation, which is a recently proposed feature selection approach, is used to choose representative bands from the noise-reduced data.  ...  In this paper, we propose a new strategy to automatically select bands without manual band removal. Firstly, wavelet shrinkage is applied to denoise the spatial images of the whole data cube.  ...  Recently, we have introduced a new clustering algorithm, based on affinity propagation (AP), for hyperspectral band selection [38] .  ... 
doi:10.1109/jstars.2012.2187434 fatcat:maknuuaum5clzhfklsckveizxm

Band selection of hyperspectral-image based weighted indipendent component analysis

Mojtaba Amini Omam, Farah Torkamani-Azar
2010 Optical Review  
Huge amounts of data in hyperspectral images have been caused to represent approaches for the band selection of these images.  ...  Applying a negentropy function to weight bands is a novel idea that leads to the selection of bands with minimum mutual information (MI) and besides maximum entropy, with respected to the bands selected  ...  Conclusions In this paper, we propose a new method based on independent component analysis (ICA) for band selection in hyperspectral images.  ... 
doi:10.1007/s10043-010-0067-7 fatcat:u5svcammnvasznvjts42v7ebpu

Dimensionality Reduction and Channel Selection of Motor Imagery Electroencephalographic Data

Muhammad Naeem, Clemens Brunner, Gert Pfurtscheller
2009 Computational Intelligence and Neuroscience  
An automated selection of ICA components based on a variance criterion was also carried out. Only 8 components chosen this way performed better (63.1%) than visually selected components.  ...  In contrast, 6 ICA components selected on the basis of visual inspection performed comparably (61.9%) to the full range of 22 components (63.9%).  ...  However, spatial filtering algorithms based on ICA can reduce the dimensionality of the data by visual selection of the components [12] on the basis of timefrequency maps [13] and scalp maps [14]  ... 
doi:10.1155/2009/537504 pmid:19536346 pmcid:PMC2695957 fatcat:d3r2cmi7qrg3rjn22j23igldtq

Adaptive color space transform using independent component analysis

Esteban Vera, Sergio Torres, Jaakko T. Astola, Karen O. Egiazarian, Edward R. Dougherty
2007 Image Processing: Algorithms and Systems V  
The new color space is used for a large set of test color images, and it is compared to traditional color space transforms, where we can clearly visualize its vast potential as a promising tool for segmentation  ...  The result is a linear and reversible color space transform that provides three new coordinate axes where the projected data is as much as statistically independent as possible, and therefore highly uncorrelated  ...  Acknowledgments The authors thank the Chilean Council for Science and Technology, CONICYT, through its project ALMA-CONICYT 30105006 and a doctorate studies scholarship, and also thank the Department of  ... 
doi:10.1117/12.705004 dblp:conf/ipas/VeraT07 fatcat:s7uzgvgu3zfgbkgtjbcjxvcdta

Simultaneous Estimation of Nongaussian Components and Their Correlation Structure

Hiroaki Sasaki, Michael U. Gutmann, Hayaru Shouno, Aapo Hyvärinen
2017 Neural Computation  
Applications on simulated complex cells with natural image input, as well as spectrograms of natural audio data, show that the method finds new kinds of dependencies between the components.  ...  To reduce the computational cost, in this experiment, at every repeat of steps 1 and 2 in algorithm 1, we randomly selected 30,000 data points from T = 100,000 data points to be used for estimation.  ...  As in the previous section, to reduce the computational cost, we randomly selected a subset of data points from the entire data points at every repeat of the two steps in algorithm 1.  ... 
doi:10.1162/neco_a_01006 pmid:28777730 fatcat:kq6fjjlij5ellpmthcjyxmndp4

The Fisher-Markov Selector: Fast Selecting Maximally Separable Feature Subset for Multiclass Classification with Applications to High-Dimensional Data

Qiang Cheng, Hongbo Zhou, Jie Cheng
2011 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Selecting features for multi-class classification is a critically important task for pattern recognition and machine learning applications.  ...  Toward the selection of the globally optimal subset of features efficiently, we introduce a new selector -which we call the Fisher-Markov selector -to identify those features that are the most useful in  ...  For these data, feature selection is often used for dimensionality reduction.  ... 
doi:10.1109/tpami.2010.195 pmid:21493968 fatcat:3djxd6onlfb43dpf4vbwee7szq

Fusion of Remote Sensing Images Using Improved ICA Mergers Based on Wavelet Decomposition

Fengrui Chen, Fen Qin, Guangxiong Peng, Shiqiang Chen
2012 Procedia Engineering  
Firstly, a convenient way which uses negentropy to measure the nongaussianity of IC is presented to select main body independent component (MBIC); secondly, in order to avoid too much spatial information  ...  In former studies, we found the fusion method based on independent component analysis (ICA) could solve this problem effectively, and attain a better balance between spectral and spatial information of  ...  For this reason, an improved ICA fusion method is presented in current study. First, a new strategy which uses negentropy to measure the nongaussianity of IC is introduced to select MBIC.  ... 
doi:10.1016/j.proeng.2012.01.418 fatcat:x6t7om3ifvbtdg7z7fr7g26ylm

Multivariate Analysis and Machine Learning for Ripeness Classification of Cape Gooseberry Fruits

Miguel De-la-Torre, Omar Zatarain, Himer Avila-George, Mirna Muñoz, Jimy Oblitas, Russel Lozada, Jezreel Mejía, Wilson Castro
2019 Processes  
component analysis, eigenvector centrality feature selection, and multi-cluster feature selection.)  ...  This paper explores five multivariate techniques for information fusion on sorting the visual ripeness of Cape gooseberry fruits (principal component analysis, linear discriminant analysis, independent  ...  The algorithms for feature extraction and selection, find a new set X = {x 1 , x 2 , ..., x k }, where x i ∈ R k , and k ≤ d is the new dimension of the feature vectors.  ... 
doi:10.3390/pr7120928 fatcat:23u3ilxo6zaqnfd2z4gaaydi6m

Independent Component Analysis in ECG Signal Processing [chapter]

Jarno M.A., Jari J.
2012 Advances in Electrocardiograms - Methods and Analysis  
FastICA is an iterative numerical algorithm, which has been developed by Hyvärinen (1999) , giving also instructions for parameter selection.  ...  On the other hand, failing ICA with the specific input signals and using one algorithm with certain parameters, may not mean that the data at hand was unfit for ICA in general.  ...  The five sections of this volume, Cardiac Anatomy, ECG Technique, ECG Features, Heart Rate Variability and ECG Data Management, provide comprehensive reviews of advancements in the technical and analytical  ... 
doi:10.5772/22719 fatcat:5xmxvgoeg5a4bfzbpjd3w2th7a

Defect detection in textile fabric images using subband domain subspace analysis

A. Serdaroglu, A. Ertuzun, A. Ercil
2007 Pattern Recognition and Image Analysis  
In this work, a new model that combines the concepts of wavelet transformation and subspace analysis tools, like Independent Component Analysis, Topographic Independent Component Analysis, and Independent  ...  Subspace Analysis, is developed for the purpose of defect detection in textile images.  ...  Conclusions In this work, a new and efficient defect detection algorithm which combines concepts of sub-band domain and subspace analysis methods is developed.  ... 
doi:10.1134/s105466180704027x fatcat:x2pmgf7n2vfanckbyi42pb25ie

Gradient Hyperalignment for multi-subject fMRI data alignment [article]

Tonglin Xu, Muhammad Yousefnezhad, Daoqiang Zhang
2018 arXiv   pre-print
Multi-subject fMRI data analysis is an interesting and challenging problem in human brain decoding studies.  ...  This paper proposes Gradient Hyperalignment (Gradient-HA) as a gradient-based functional alignment method that is suitable for multi-subject fMRI datasets with large amounts of samples and voxels.  ...  This algorithm can solve the problem of fMRI data with multiple subjects, a large number of samples and plenty of voxels.  ... 
arXiv:1807.02612v1 fatcat:di67gtcvzbbe3jbbqkphgcw2wu

Some Statistical Methods for Random Process Data from Seismology and Neurophysiology [chapter]

David R. Brillinger
2011 Selected Works of David Brillinger  
Examples are presented of statistical techniques for the analysis of random process data and of their uses in the substantive fields of seismology and neurophysiology.  ...  The problems addressed include frequency estimation for decaying cosinusoids, signal estimation, association measurement, causal connection assessment, estimation of speed and direction and structural  ...  I can say that these lectures would never have come about, but for the help and encouragement that my collaborators B. A. Bolt and  ... 
doi:10.1007/978-1-4614-1344-8_8 fatcat:jcxskipblfcnrieizk3vxqkuri

Ultrasonic image classification based on support vector machine with two independent component features

Weishi Chen, Tiejun Liu, Baofa Wang
2011 Computers and Mathematics with Applications  
Finally, some new technique is suggested for algorithm improvement in the future.  ...  After training of selected samples, a support vector machine (SVM) classifier which combined the two ICA representations is established for recognition, and a good performance is given for testing data  ...  The authors wish to acknowledge the TEKNOVA Medical Systems Ltd. and Beijing Hospital of Integrated Traditional Chinese and Western Medicine for their help in ultrasonic image collection.  ... 
doi:10.1016/j.camwa.2011.06.051 fatcat:7ucenfa3uzgtngu3qtac64f43e

Image Denoising with Directional Bases

Heechan Park, Graham R Martin, Zhen Yao
2007 Proceedings of IEEE international conference on image processing  
We then present a combined approach that benefits from the computational efficiency of the MFT and the data adaptiveness of ICA.  ...  One involves the Multi-resolution Fourier Transform (MFT) facilitated with a multi-directional selective filter.  ...  linear mixing matrix and the source (latent) data respectively, which are independent and nongaussian.  ... 
doi:10.1109/icip.2007.4378951 dblp:conf/icip/ParkMY07 fatcat:g4xhm6gln5cpnofou3qlcxl2h4
« Previous Showing results 1 — 15 out of 309 results