2004
DOI: 10.1016/s0262-8856(03)00136-7
|View full text |Cite
|
Sign up to set email alerts
|

Non-parametric and unsupervised Bayesian classification with Bootstrap sampling

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
10
0

Year Published

2004
2004
2016
2016

Publication Types

Select...
5
2

Relationship

2
5

Authors

Journals

citations
Cited by 14 publications
(10 citation statements)
references
References 26 publications
0
10
0
Order By: Relevance
“…• In the region technique, segmentation using the BEM algorithm (see Section 2.1) is applied. First, the optimal size of the bootstrap sample is determined and then this sample is drawn using the replacement of a representative bootstrap sample from the totality of the image (Delleji, Zribi, and Hamida 2007;Zribi 2004). Therefore, the k-means clustering method was used to initialize the number of classes and their parameters for the input image (Kehtarnavaz and Nakamura 1998;Masson and Pieczynski 1993).…”
Section: Computation Process Of Correlation Mapmentioning
confidence: 99%
“…• In the region technique, segmentation using the BEM algorithm (see Section 2.1) is applied. First, the optimal size of the bootstrap sample is determined and then this sample is drawn using the replacement of a representative bootstrap sample from the totality of the image (Delleji, Zribi, and Hamida 2007;Zribi 2004). Therefore, the k-means clustering method was used to initialize the number of classes and their parameters for the input image (Kehtarnavaz and Nakamura 1998;Masson and Pieczynski 1993).…”
Section: Computation Process Of Correlation Mapmentioning
confidence: 99%
“…A large variety of different segmentation algorithms have been developed [35][36][37]. An unsupervised Bayesian segmentation based on Bootstrapping NEM algorithm was proposed in our previous research [23]. These region characteristics favor the NEM fusion algorithm performed on each region in the next step.…”
Section: Unsupervised Bayesian Segmentation Image By Bootstrapping Nementioning
confidence: 99%
“…Thus, the parameters' estimation would be made with only the use of a small number of representative samples instead of all correlated pixels in the real image. Zribi [23] introduced the same approach to the non-parametric Expectation-Maximization (NEM) algorithm [24]. The non-parametric aspect comes from the use of the orthogonal probability density function (pdf) estimation, which could be simply reduced to the estimation of the first Fourier Coefficients (FC's) of the pdf with respect to a given orthogonal basis [25,26].…”
Section: Introductionmentioning
confidence: 99%
“…Many researchers have applied the Bootstrap approach in pattern classification [16,18,23] and in image segmentation [15]. In this section, we present the application of BSEM algorithm firstly for unsupervised image segmentation, and secondly for image fusion process.…”
Section: Bsem Algorithm Applicationmentioning
confidence: 99%
“…In fact, we tried to combine the Bootstrap approach [16][17][18] with the Stochastic EM (SEM) algorithm [19,20].…”
Section: Introductionmentioning
confidence: 99%