1. Extension of REBMIX algorithm to von Mises parametric family for modeling joint distribution of wind speed and directionX. W. Ye, P. S. Xi, Marko Nagode, 2019, original scientific article Keywords: structural health monitoring, wind field characteristics, joint probability density function, mixture estimation, REBMIX algorithm
Full text (file, 2,59 MB) This document has more files! More... |
2. Optimizing the estimation of a histogram-bin width—application to the multivariate mixture-model estimationBranislav Panić, Jernej Klemenc, Marko Nagode, 2020, original scientific article Keywords: histogram, integer optimization, parameter estimation, mixture model, EM, REBMIX
Full text (file, 14,59 MB) This document has more files! More... |
3. Improved initialization of the EM algorithm for mixture model parameter estimationBranislav Panić, Jernej Klemenc, Marko Nagode, 2020, original scientific article Keywords: mixture model, parameter estimation, EM algorithm, REBMIX algorithm, density estimation, clustering, image segmentation
Full text (file, 9,24 MB) This document has more files! More... |
|
5. Gaussian mixture model based classification revisitedBranislav Panić, Jernej Klemenc, Marko Nagode, 2020, original scientific article Keywords: Gaussian mixture models, classification, bearing fault estimation, parameter estimation, performance of classification methods
Full text (file, 1,01 MB) This document has more files! More... |
6. Gumbel mixture modelling for multiple failure dataMarko Nagode, Simon Oman, Jernej Klemenc, Branislav Panić, 2023, original scientific article Keywords: Gumbel mixture model, parameter estimation, system safety, failure modelling
Full text (file, 1,35 MB) This document has more files! More... |
7. On methods for merging mixture model components suitable for unsupervised image segmentation tasksBranislav Panić, Marko Nagode, Jernej Klemenc, Simon Oman, 2022, original scientific article Keywords: mixture models, parameter estimation, clustering, unsupervised image segmentation
Full text (file, 3,79 MB) This document has more files! More... |
