Filters








480 Hits in 7.6 sec

Initializing nonnegative matrix factorization using the successive projection algorithm for multi-parametric medical image segmentation

Nicolas Sauwen, Marjan Acou, Halandur Nagaraja Bharath, Diana Maria Sima, Jelle Veraart, Frederik Maes, Uwe Himmelreich, Eric Achten, Sabine Van Huffel
2016 The European Symposium on Artificial Neural Networks  
This study proposes the Successive Projection Algorithm (SPA) as a feasible NMF initialization method. SPA is applied to a multi-parametric MRI dataset for automated NMF brain tumor segmentation.  ...  As nonnegative matrix factorization (NMF) represents a nonconvex problem, the quality of its solution will depend on the initialization of the factor matrices.  ...  Acknowledgements This work has been funded by: Flemish Government FWO project G. 0869  ... 
dblp:conf/esann/SauwenABSVMHAH16 fatcat:z5fs67txxfe2lmg6yrlgyja6nu

Variational Viewpoint of the Quadratic Markov Measure Field Models: Theory and Algorithms

M. Rivera, O. Dalmau
2012 IEEE Transactions on Image Processing  
We present a framework for image segmentation based on quadratic programing; i.e. by the minimization of a quadratic regularized energy linearly constrained.  ...  We demonstrate the proposal capabilities by experiments and numerical comparisons with interactive two-class segmentation as well as in the simultaneous estimation of segmentation and (parametric and non-parametric  ...  The segmentation were computed with the multi-class EC-QMMF algorithm using color histograms, see section 5.1. for faster algorithms is, indeed, an active research topic.  ... 
doi:10.1109/tip.2011.2168409 pmid:21926023 fatcat:opafehiq3vgtdef75fh4avfisu

Segmentation of Infant Brain Using Nonnegative Matrix Factorization

Norah Saleh Alghamdi, Fatma Taher, Heba Kandil, Ahmed Sharafeldeen, Ahmed Elnakib, Ahmed Soliman, Yaser ElNakieb, Ali Mahmoud, Mohammed Ghazal, Ayman El-Baz
2022 Applied Sciences  
The visual appearance is modeled with an empirical distribution of the probability of the DTI features, fused by their nonnegative matrix factorization (NMF) and allocation to data clusters.  ...  Projecting an initial high-dimensional feature space onto a low-dimensional space of the significant fused features with the NMF allows for better separation of the goal structure and its background.  ...  The initial segmentation for the various brain structures was performed using patching, along with a ground truth-aligned multi-modal image library.  ... 
doi:10.3390/app12115377 fatcat:nqcadvr4ufgchlwqs6i4xd37xu

Hyperspectral image segmentation, deblurring, and spectral analysis for material identification

Fang Li, Michael K. Ng, Robert Plemmons, Sudhakar Prasad, Qiang Zhang, Zia-ur Rahman, Stephen E. Reichenbach, Mark A. Neifeld
2010 Visual Information Processing XIX  
Since multi-spectral or hyperspectral imagery is generally low resolution, it is possible for pixels in the image to contain several materials.  ...  An important aspect of spectral image analysis is identification of materials present in the object or scene being imaged.  ...  This approach preserves physical characteristics of the data such as nonnegativity and is a natural extension of nonnegative least squares approximate nonnegative matrix factorization. 2 However, the  ... 
doi:10.1117/12.850121 dblp:conf/spieVIP/LiNPPZ10 fatcat:lsp7ib4zbnehvckzz3rg2a4sua

A Review on Prostate Cancer Detection using Deep Learning Techniques

Narmatha C, Department of Computer Science, Faculty of Computers and Information Technology, University of Tabuk, Saudi Arabia., Surendra Prasad M, Annapoorana Medical College and Hospitals. Salem, Tamil Nadu, India
2020 Journal of Computational Science and Intelligent Technologies  
For segmentation and classification, deep learning algorithms, specifically convolutional networks, have quickly become an optional technique for medical image analysis.  ...  Prostate segmentation and classification is a difficult process, and the difficulties fundamentally vary with one imaging methodology then onto the next.  ...  These three types of features were combined utilizing a nonnegative matrix factorization (NMF) method.  ... 
doi:10.53409/mnaa.jcsit20201204 fatcat:ek3dwtjutvgnzkrd23sja3xdq4

Unsupervised contour representation and estimation using B-splines and a minimum description length criterion

M.A.T. Figueiredo, J.M.N. Leitao, A.K. Jain
2000 IEEE Transactions on Image Processing  
The experiments reported in the paper, performed on synthetic and real (medical) images, confirm the adequacy and good performance of the approach.  ...  The result is an unsupervised parametric deformable contour: it adapts its degree of smoothness/complexity (number of control points) and it also estimates the observation (image) model parameters.  ...  The key factor controlling the success of a given initial contour is the resulting initial parameter estimate; more precisely, the first estimate of must be such that Fig. 15 .  ... 
doi:10.1109/83.846249 pmid:18255477 fatcat:bcfvkv52hnfyvcpfucj65qc2ae

Bayesian Nonparametric Approaches to Abnormality Detection in Video Surveillance

Vu Nguyen, Dinh Phung, Duc-Son Pham, Svetha Venkatesh
2015 Annals of Data Science  
In particular, we employ the Infinite Hidden Markov Model and Bayesian Nonparametric Factor Analysis for stream data segmentation and pattern discovery.  ...  However, the crucial challenge in stream data segmentation and hidden patterns discovery are the number of coherent segments in surveillance stream and the number of traffic patterns are unknown and hard  ...  Using a part-based decomposition approach, a straightforward approach is to use nonnegative matrix factorization (NNMF) of [25] which factorizes X into X ≈ W H, (1) where W and H are nonnegative matrices  ... 
doi:10.1007/s40745-015-0030-3 fatcat:56wmpggdgraktipdvg5solnnp4

Theta phase coding in a network model of the entorhinal cortex layer II with entorhinal-hippocampal loop connections

Jun Igarashi, Hatsuo Hayashi, Katsumi Tateno
2006 Cognitive Neurodynamics  
NMF seeks a decomposition of a nonnegative data matrix into a product of two factor matrices (basis matrix and encoding matrix) such that all factor matrices are forced to be nonnegative.  ...  Among those, nonnegative matrix factorization (NMF) has recently drawn extensive attention, since promising results were reported in handling nonnegative data such as document, image data, spectrograms  ...  The proposed algorithms can be implemented in pixel-parallel, and they use only the neighboring pixel states, they are suitable for the VLSI implementation.  ... 
doi:10.1007/s11571-006-9003-8 pmid:19003510 pmcid:PMC2267667 fatcat:avihqfr6a5e7tm43lythi27n3a

Quadratic Markovian Probability Fields for Image Binary Segmentation

Mariano Rivera, Pedro P. Mayorga
2007 2007 IEEE 11th International Conference on Computer Vision  
We compare the algorithms using cross-validation procedures and a simplex decent algorithm for learning the parameter set.  ...  In this work we present a new Markov Random Field model for image binary segmentation that computes the probability that each pixel belongs to a given class.  ...  Such a factor is also changed for estimating the covariance matrix of multivariated Gaussian models. Figure 4 shows the pair of images used this experiment.  ... 
doi:10.1109/iccv.2007.4409119 dblp:conf/iccv/RiveraM07 fatcat:rc4ixqlsyzgzdi6q7ru4go5qly

Generative-Discriminative Basis Learning for Medical Imaging

N. K. Batmanghelich, B. Taskar, C. Davatzikos
2012 IEEE Transactions on Medical Imaging  
Effectiveness of the proposed algorithm for SSL is evaluated with both benchmark and medical imaging datasets.  ...  For evaluation of the SSL setting in medical datasets, we use images of subjects with Mild Cognitive Impairment (MCI), which is believed to be a precursor to Alzheimer's disease (AD), as unlabeled data  ...  The authors would like to thank the anonymous reviewers for their valuable comments and suggestions to improve the quality of the paper.  ... 
doi:10.1109/tmi.2011.2162961 pmid:21791408 pmcid:PMC3402718 fatcat:bjpkbh3hbnd4zemresxl7dewvm

Low-Rank Modeling and Its Applications in Image Analysis [article]

Xiaowei Zhou, Can Yang, Hongyu Zhao, Weichuan Yu
2014 arXiv   pre-print
Then, we summarize the models and algorithms for low-rank matrix recovery and illustrate their advantages and limitations with numerical experiments.  ...  In this paper, we review the recent advance of low-rank modeling, the state-of-the-art algorithms, and related applications in image analysis.  ...  ACKNOWLEDGMENTS The authors thank Michael Klein for carefully proofreading the manuscript.  ... 
arXiv:1401.3409v3 fatcat:tfniuthcavehpfoervf2p4ek6a

Low-Rank Modeling and Its Applications in Image Analysis

Xiaowei Zhou, Can Yang, Hongyu Zhao, Weichuan Yu
2014 ACM Computing Surveys  
Then, we summarize the models and algorithms for low-rank matrix recovery and illustrate their advantages and limitations with numerical experiments.  ...  In this paper, we review the recent advance of low-rank modeling, the stateof-the-art algorithms, and related applications in image analysis.  ...  ACKNOWLEDGMENTS The authors thank Michael Klein for carefully proofreading the manuscript.  ... 
doi:10.1145/2674559 fatcat:7ph4bazvuffwrc7teaoqdit3ju

ÉCole De Physique Des Houches [chapter]

2007 Les Houches  
activity in ideal surroundings for skiing, hiking or mountaineering.  ...  Established in 1951, the Physics School is located in a group of chalets surrounded by meadows and woods, at an altitude of 1150 m facing the Mont-Blanc range -a very favourable environment for intellectual  ...  The authors would like to thank Dr. Naz Chaudary and Dr. Richard Hill for providing the ME180 tumor bearing mice and Mr. Michael Dunne for providing the H520 tumor bearing mice.  ... 
doi:10.1016/s0924-8099(13)60020-7 fatcat:jydcwidbbnf3nppqp4x3pyu6ga

École de Physique des Houches [chapter]

2008 Les Houches  
activity in ideal surroundings for skiing, hiking or mountaineering.  ...  Established in 1951, the Physics School is located in a group of chalets surrounded by meadows and woods, at an altitude of 1150 m facing the Mont-Blanc range -a very favourable environment for intellectual  ...  The authors would like to thank Dr. Naz Chaudary and Dr. Richard Hill for providing the ME180 tumor bearing mice and Mr. Michael Dunne for providing the H520 tumor bearing mice.  ... 
doi:10.1016/s0924-8099(13)60004-9 fatcat:nxiezedqabbbtnfqonvw6c3doi

École de Physique des Houches [chapter]

2006 Les Houches  
activity in ideal surroundings for skiing, hiking or mountaineering.  ...  Established in 1951, the Physics School is located in a group of chalets surrounded by meadows and woods, at an altitude of 1150 m facing the Mont-Blanc range -a very favourable environment for intellectual  ...  The authors would like to thank Dr. Naz Chaudary and Dr. Richard Hill for providing the ME180 tumor bearing mice and Mr. Michael Dunne for providing the H520 tumor bearing mice.  ... 
doi:10.1016/s0924-8099(06)80084-3 fatcat:6w7diik2k5af7cpowh4xz3nsky
« Previous Showing results 1 — 15 out of 480 results