MULTI-REGULARIZATION PARAMETERS ESTIMATION FOR GAUSSIAN MIXTURE CLASSIFIER BASED ON MDL PRINCIPLE

In order to improve the accuracy of the image segmentation in video surveillance sequences and to overcome the limits of the traditional clustering algorithms that can not accurately model the image data sets which Contains noise data, the paper presents an automatic and accurate video image segmentation algorithm, according to the spatial properties, which uses the Gaussian mixture models to segment the image. But the expectation-maximization algorithm is very sensitive to initial values, and easy to fall into local optimums, so the paper presents a differential evolution-based parameters estimation for Gaussian mixture models. The experiment result shows that the segmentation accuracy has been improved greatly than by the traditional segmentation algorithms.

Download Full-text

LS-EM Algorithm of Parameters Estimation for Gaussian Mixture Autoregressive Model

The Proceedings of the Multiconference on "Computational Engineering in Systems Applications" ◽

10.1109/cesa.2006.4281613 ◽

2006 ◽

Cited By ~ 1

Author(s):

Wang Ping-bo ◽

Cai Zhi-ming

Keyword(s):

Em Algorithm ◽

Autoregressive Model ◽

Gaussian Mixture ◽

Parameters Estimation

Download Full-text

Mixed Platoon Flow Dispersion Model Based on Speed-Truncated Gaussian Mixture Distribution

Journal of Applied Mathematics ◽

10.1155/2013/480965 ◽

2013 ◽

Vol 2013 ◽

pp. 1-9 ◽

Cited By ~ 3

Author(s):

Weitiao Wu ◽

Wenzhou Jin ◽

Luou Shen

Keyword(s):

Model Comparison ◽

Dispersion Model ◽

Mixture Distribution ◽

Flow Distribution ◽

Gaussian Mixture ◽

Parameters Estimation ◽

The Road ◽

Proposed Model ◽

Truncated Gaussian ◽

Gaussian Mixture Distribution

A mixed traffic flow feature is presented on urban arterials in China due to a large amount of buses. Based on field data, a macroscopic mixed platoon flow dispersion model (MPFDM) was proposed to simulate the platoon dispersion process along the road section between two adjacent intersections from the flow view. More close to field observation, truncated Gaussian mixture distribution was adopted as the speed density distribution for mixed platoon. Expectation maximum (EM) algorithm was used for parameters estimation. The relationship between the arriving flow distribution at downstream intersection and the departing flow distribution at upstream intersection was investigated using the proposed model. Comparison analysis using virtual flow data was performed between the Robertson model and the MPFDM. The results confirmed the validity of the proposed model.

Download Full-text

Optical Coherence Tomography Noise Reduction Using Anisotropic Local Bivariate Gaussian Mixture Prior in 3D Complex Wavelet Domain

International Journal of Biomedical Imaging ◽

10.1155/2013/417491 ◽

2013 ◽

Vol 2013 ◽

pp. 1-23 ◽

Cited By ~ 14

Author(s):

Hossein Rabbani ◽

Milan Sonka ◽

Michael D. Abramoff

Keyword(s):

A Priori ◽

Gaussian Mixture ◽

Parameters Estimation ◽

Wavelet Domain ◽

Free Data ◽

Mixture Prior ◽

Mmse Estimator ◽

Complex Wavelet ◽

Heavy Tailed ◽

Local Parameters

In this paper, MMSE estimator is employed for noise-free 3D OCT data recovery in 3D complex wavelet domain. Since the proposed distribution for noise-free data plays a key role in the performance of MMSE estimator, a priori distribution for the pdf of noise-free 3D complex wavelet coefficients is proposed which is able to model the main statistical properties of wavelets. We model the coefficients with a mixture of two bivariate Gaussian pdfs with local parameters which are able to capture the heavy-tailed property and inter- and intrascale dependencies of coefficients. In addition, based on the special structure of OCT images, we use an anisotropic windowing procedure for local parameters estimation that results in visual quality improvement. On this base, several OCT despeckling algorithms are obtained based on using Gaussian/two-sided Rayleigh noise distribution and homomorphic/nonhomomorphic model. In order to evaluate the performance of the proposed algorithm, we use 156 selected ROIs from 650 × 512 × 128 OCT dataset in the presence of wet AMD pathology. Our simulations show that the best MMSE estimator using local bivariate mixture prior is for the nonhomomorphic model in the presence of Gaussian noise which results in an improvement of 7.8 ± 1.7 in CNR.

Download Full-text