Unsupervised selection and estimation of finite mixture models

Figueiredo, Mário A. T.; Jain, Anshul

doi:10.1109/icpr.2000.906023

Cited by 40 publications

(44 citation statements)

References 17 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In contrast, we determine the appropriate number of components and the parameters of each component simultaneously, by minimizing the entropy of the component cardinalities while maximizing the log-likelihood function. Here we would like to note that more recently, there have been attempts at using the above-mentioned criteria more efficiently and even combining an MDL-type criterion with the log likelihood [18][19][20].…”

Section: Gaussian Mixture Decompositionmentioning

confidence: 98%

Categorization of Image Databases for Efficient Retrieval Using Robust Mixture Decomposition

Medasani

Krishnapuram

2001

Computer Vision and Image Understanding

View full text Add to dashboard Cite

Section: Gaussian Mixture Decompositionmentioning

confidence: 98%

Categorization of Image Databases for Efficient Retrieval Using Robust Mixture Decomposition

Medasani

Krishnapuram

2001

Computer Vision and Image Understanding

View full text Add to dashboard Cite

“…There exist many modifications to his basic procedure such as changing the complexity of the model via self-annealing algorithms [22]. EM guarantees a monotonically nondecreasing likelihood [23] although its ability to find a local maximum depends on parameter initialization.…”

Section: M-mentioning

confidence: 99%

A portable electronic nose based on embedded PC technology and GNU/Linux: hardware, software and applications

et al. 2002

View full text Add to dashboard Cite

Abstract-This paper describes a portable electronic nose based on embedded PC technology. The instrument combines a small footprint with the versatility offered by embedded technology in terms of software development and digital communications services. A summary of the proposed hardware and software solutions is provided with an emphasis on data processing. Data evaluation procedures available in the instrument include automatic feature selection by means of SFFS, feature extraction with linear discriminant analysis (LDA) and principal component analysis (PCA), multi-component analysis with partial least squares (PLS) and classification through -NN and Gaussian mixture models. In terms of instrumentation, the instrument makes use of temperature modulation to improve the selectivity of commercial metal oxide gas sensors. Field applications of the instrument, including experimental results, are also presented.

show abstract

“…In our original approach we fit a mixture of Gaussians to the distribution of leaf nodes w f using the approach of Figueiredo [18] and the final segmentation was found by the Gaussian that gave the highest probability.…”

Section: Bottom-up Merging Of Regionsmentioning

confidence: 99%

Texture Segmentation by Contractive Decomposition and Planar Grouping

Dahl

Bogunovich

Shokoufandeh

2009

Graph-Based Representations in Pattern Recognition

View full text Add to dashboard Cite

Abstract. Image segmentation has long been an important problem in the computer vision community. In our recent work we have addressed the problem of texture segmentation, where we combined top-down and bottom-up views of the image into a unified procedure. In this paper we extend our work by proposing a modified procedure which makes use of graphs of image regions. In the topdown procedure a quadtree of image region descriptors is obtained in which a novel affine contractive transformation based on neighboring regions is used to update descriptors and determine stable segments. In the bottom-up procedure we form a planar graph on the resulting stable segments, where edges are present between vertices representing neighboring image regions. We then use a vertex merging technique to obtain the final segmentation. We verify the effectiveness of this procedure by demonstrating results which compare well to other recent techniques.

show abstract

Unsupervised selection and estimation of finite mixture models

Abstract: We

Cited by 40 publications

References 17 publications

Categorization of Image Databases for Efficient Retrieval Using Robust Mixture Decomposition

Categorization of Image Databases for Efficient Retrieval Using Robust Mixture Decomposition

A portable electronic nose based on embedded PC technology and GNU/Linux: hardware, software and applications

Texture Segmentation by Contractive Decomposition and Planar Grouping

Contact Info

Product

Resources

About