Filters








1,497 Hits in 4.4 sec

Image Similarity Based on Hierarchies of ICA Mixtures [chapter]

Arturo Serrano, Addisson Salazar, Jorge Igual, Luis Vergara
Independent Component Analysis and Signal Separation  
The procedure is applied to merge similar patches on a natural image, to group different images of an object, and to create hierarchical levels of clustering from images of different objects.  ...  The hierarchy algorithm composes an agglomerative (bottom-up) clustering from the estimated parameters (basis vectors and bias terms) of the ICA mixture.  ...  (Right) Hierarchical representation of the zones of the image based on basis functions similarity.  ... 
doi:10.1007/978-3-540-74494-8_98 dblp:conf/ica/SerranoSIV07 fatcat:l6ablkxlz5hffk3yisykvy64zy

Capturing nonlinear dependencies in natural images using ICA and mixture of Laplacian distribution

Hyun-Jin Park, Te-Won Lee
2006 Neurocomputing  
The model makes use of lower level linear ICA representation and a subsequent mixture of Laplacian distribution for learning the nonlinear dependencies in an image.  ...  The model makes use of lower level linear ICA representation and a subsequent mixture of Laplacian distribution for learning the nonlinear dependencies in an image.  ...  First, we can exploit the regularity of the image segmentation result to learn higher-level structures by building additional hierarchies on the current model.  ... 
doi:10.1016/j.neucom.2005.12.026 fatcat:jqvmqzxz6nfo5j566tvmhxfrwy

Is Early Vision Optimized for Extracting Higher-order Dependencies?

Yan Karklin, Michael S. Lewicki
2005 Neural Information Processing Systems  
Here we examine the optimal lower-level representations derived in the context of a hierarchical model and find that the resulting representations are strikingly different from those based on linear models  ...  Our work unifies several related approaches and observations about natural image structure and suggests that hierarchical models might yield better representations of image structure throughout the hierarchy  ...  It is interesting that the learned representations are similar to the results obtained when ICA or sparse coding is applied to whitened images (i.e. with a flattened power spectrum).  ... 
dblp:conf/nips/KarklinL05 fatcat:363bjjcjtzcy5kre6d76mln23a

Modeling Nonlinear Dependencies in Natural Images using Mixture of Laplacian Distribution

Hyun-Jin Park, Te-Won Lee
2004 Neural Information Processing Systems  
This method is an extension of the linear Independent Component Analysis (ICA) method by building a hierarchical model based on ICA and mixture of Laplacian distribution.  ...  We propose a new method for capturing nonlinear dependencies in images of natural scenes.  ...  model based on ICA and a mixture of Laplacian distribution.  ... 
dblp:conf/nips/ParkL04 fatcat:zqo3uchprfey5lhdm44oqtwck4

Correspondence of the brain's functional architecture during activation and rest

S. M. Smith, P. T. Fox, K. L. Miller, D. C. Glahn, P. M. Fox, C. E. Mackay, N. Filippini, K. E. Watkins, R. Toro, A. R. Laird, C. F. Beckmann
2009 Proceedings of the National Academy of Sciences of the United States of America  
In this work, we identify the major explicit activation networks by carrying out an image-based activation network analysis of thousands of separate activation maps derived from the BrainMap database of  ...  Independently, we extract the major covarying networks in the resting brain, as imaged with functional magnetic resonance imaging in 36 subjects at rest.  ...  ., and A.R.L. were supported, as part of the Human Brain Project, by National Institute of Mental Health Grant R01-MH074457-01A1 (P.T.F., principal investigator).  ... 
doi:10.1073/pnas.0905267106 pmid:19620724 pmcid:PMC2722273 fatcat:6btlwl3gbvfkjojcyymhcejwde

Assignment of Multiplicative Mixtures in Natural Images

Odelia Schwartz, Terrence J. Sejnowski, Peter Dayan
2004 Neural Information Processing Systems  
We demonstrate the efficacy of the approach on both synthetic and image data. * Mixer variables are also called mutlipliers, but are unrelated to the scales of a wavelet.  ...  In the analysis of natural images, Gaussian scale mixtures (GSM) have been used to account for the statistics of filter responses, and to inspire hierarchical cortical representational learning schemes  ...  based on cortical development) towards the coordinated statistical structure of the wavelet components.  ... 
dblp:conf/nips/SchwartzSD04 fatcat:bobhec55dzc2ndx33tqexijvue

A Hierarchical Model for Probabilistic Independent Component Analysis of Multi-Subject fMRI Studies

Ying Guo, Li Tang
2013 Biometrics  
Existing group ICA methods generally concatenate observed fMRI data across subjects on the temporal domain and then decompose multi-subject data in a similar manner to single-subject ICA.  ...  The proposed method provides model-based estimation of brain functional networks at both the population and subject level.  ...  Giuseppe Pagnoni for providing the Zen meditation data and interpretations of the results. This work was supported by NIH grants UL1TR000454 (NCATS) and 1 U18 NS082143-01 (NINDS).  ... 
doi:10.1111/biom.12068 pmid:24033125 pmcid:PMC4130464 fatcat:k6cvz7lmvrfhviw2qnr4vkq7ma

Estimation of the Number of Endmembers in Hyperspectral Images Using Agglomerative Clustering

José Prades, Gonzalo Safont, Addisson Salazar, Luis Vergara
2020 Remote Sensing  
Based on the K-means result, a model-based agglomerative clustering is performed, which provides a hierarchy of clusterings.  ...  The algorithm is based on the assumption that a valid clustering of the image has one cluster for each different material.  ...  However, the algorithm in [23] assumes that the feature vectors of the image are drawn from an ICA mixture model [24, 25] , whose ICA parameters are jointly estimated using a maximum likelihood approach  ... 
doi:10.3390/rs12213585 fatcat:j52s7fqd2bgkffa6wsdaceiqk4

Distributed Multi-Target Detection in Sensor Networks [chapter]

Xiaoling Wang, Hairong Qi, Steve Beck
2012 Distributed Sensor Networks, Second Edition  
Take 2-D imagers as an example: through image segmentation, the targets of interest can be separated from the background and later identified using pattern classification methods.  ...  For example, a constant false-alarm rate (CFAR) detector on the acoustic signals can determine the presence of a target if the signal energy exceeds an adaptive threshold.  ...  This problem is similar to the BSS problem in signal processing and ICA is traditionally the most popular algorithm to solve it.  ... 
doi:10.1201/b12991-16 fatcat:73w7sylxurc73lcswmr6ocrvzm

Sparsity-Regularized HMAX for Visual Recognition

Xiaolin Hu, Jianwei Zhang, Jianmin Li, Bo Zhang, Gennady Cymbalyuk
2014 PLoS ONE  
Unlike most other deep learning models that explicitly address global structure of images in every layer, sparse HMAX addresses local to global structure gradually along the hierarchy by applying patch-based  ...  This model is able to learn higher-level features of objects on unlabeled training images.  ...  The parameters were as follows: 8 S1 bases of size 8|8 learned by ICA on the Kyoto images; 1024 S2 bases of size 4|4 learned by SSC on the Caltech-101 images; pooling ratio r 1 ~6.  ... 
doi:10.1371/journal.pone.0081813 pmid:24392078 pmcid:PMC3879257 fatcat:zet6adbx2rhzzfxl3fvnfshp6y

Functional Parcellation of fMRI data using multistage k-means clustering [article]

Harshit Parmar, Brian Nutter, Rodney Long, Sameer Antani, Sunanda Mitra
2022 arXiv   pre-print
One of the approaches for analysis and interpretation of resting-state fMRI data require spatially and functionally homogenous parcellation of the whole brain based on underlying temporal fluctuations.  ...  Among commonly used parcellation schemes, a tradeoff exists between intra-cluster functional similarity and alignment with anatomical regions.  ...  Javier Gonzales-Castillo, staff scientist National Institute of Mental Health NIH, for providing the 100 functional run per subject dataset  ... 
arXiv:2202.11206v1 fatcat:mtwyjrhpd5bflce4uni36npqjm

Applying FSL to the FIAC data: Model-based and model-free analysis of voice and sentence repetition priming

Christian F. Beckmann, Mark Jenkinson, Mark W. Woolrich, Timothy E.J. Behrens, David E. Flitney, Joseph T. Devlin, Stephen M. Smith
2006 Human Brain Mapping  
the use of optimal constrained HRF basis function modelling and mixture modelling inference.  ...  We also discuss the application of tools for the correction of image distortions prior to the statistical analysis and the utility of recent advances in FMRI time series modelling and inference such as  ...  based on non-spatial Gaussian/Gamma mixture modelling of the F -statistics image.  ... 
doi:10.1002/hbm.20246 pmid:16565953 pmcid:PMC2653076 fatcat:7wodnnsw5zef5i5ytmebuonvum

Soft Mixer Assignment in a Hierarchical Generative Model of Natural Scene Statistics

Odelia Schwartz, Terrence J. Sejnowski, Peter Dayan
2006 Neural Computation  
Gaussian scale mixture models offer a top-down description of signal generation that captures key bottom-up statistical characteristics of filter responses to images.  ...  We also show how our model helps interrelate a wide range of models of image statistics and cortical processing.  ...  We considered inference on both wavelet filters and ICA bases and with a number of different image sets.  ... 
doi:10.1162/neco.2006.18.11.2680 pmid:16999575 pmcid:PMC2915771 fatcat:mtgnjskpozhd5megl7tzswucti

Separation of Nonlinear Image Mixtures by Denoising Source Separation [chapter]

Mariana S. C. Almeida, Harri Valpola, Jaakko Särelä
2006 Lecture Notes in Computer Science  
Separation results on a real-world image mixture proved to be comparable to those achieved with MISEP.  ...  The denoising source separation framework is extended to nonlinear separation of image mixtures. MLP networks are used to model the nonlinear unmixing mapping.  ...  Almeida for helpful comments on this manuscript and for providing the images and related information.  ... 
doi:10.1007/11679363_2 fatcat:jxwkyzmt2vhihotwe7pnrfhovi

Deconvolution of dynamic dual photon microscopy images of cerebral microvasculature to assess the hemodynamic status of the brain

Hatef Mehrabian, Liis Lindvere, Bojana Stefanovic, Anne L. Martel, John B. Weaver, Robert C. Molthen
2011 Medical Imaging 2011: Biomedical Applications in Molecular, Structural, and Functional Imaging  
Dynamic contrast enhanced imaging of cerebral microvasculature provides information that can be used in understanding physiology of cerebral diseases.  ...  However, post-processing of the data is needed to segment the FOV and to perform Deconvolution to remove the effects of input bolus profile and the path it travels to reach the imaging window.  ...  The core idea of ICA is motivated from blind source separation problem for data model of the form where is a matrix of the N observed mixtures (frames or images), is a matrix containing the M source signals  ... 
doi:10.1117/12.878741 fatcat:gvum3b2udrbwlosmaq74pegvyq
« Previous Showing results 1 — 15 out of 1,497 results