A spatially constrained mixture model for image segmentation Export

@article{citeulike:2798973, abstract = {Gaussian mixture models (GMMs) constitute a well-known type of probabilistic neural networks. One of their many successful applications is in image segmentation, where spatially constrained mixture models have been trained using the expectation-maximization (EM) framework. In this letter, we elaborate on this method and propose a new methodology for the M-step of the EM algorithm that is based on a novel constrained optimization formulation. Numerical experiments using simulated images illustrate the superior performance of our method in terms of the attained maximum value of the objective function and segmentation accuracy compared to previous implementations of this approach.}, author = {Blekas, K. and Likas, A. and Galatsanos, N. P. and Lagaris, I. E.}, booktitle = {Neural Networks, IEEE Transactions on}, citeulike-article-id = {2798973}, citeulike-linkout-0 = {http://dx.doi.org/10.1109/TNN.2004.841773}, citeulike-linkout-1 = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1402510}, doi = {10.1109/TNN.2004.841773}, journal = {Neural Networks, IEEE Transactions on}, keywords = {constrained, gaussian, gmm, mixture, segmentation, spatially}, number = {2}, pages = {494--498}, posted-at = {2008-05-14 15:21:23}, priority = {2}, title = {A spatially constrained mixture model for image segmentation}, url = {http://dx.doi.org/10.1109/TNN.2004.841773}, volume = {16}, year = {2005} }

TY - JOUR ID - citeulike:2798973 L3 - citeulike-article-id:2798973 N2 - Gaussian mixture models (GMMs) constitute a well-known type of probabilistic neural networks. One of their many successful applications is in image segmentation, where spatially constrained mixture models have been trained using the expectation-maximization (EM) framework. In this letter, we elaborate on this method and propose a new methodology for the M-step of the EM algorithm that is based on a novel constrained optimization formulation. Numerical experiments using simulated images illustrate the superior performance of our method in terms of the attained maximum value of the objective function and segmentation accuracy compared to previous implementations of this approach. IS - 2 JF - Neural Networks, IEEE Transactions on EP - 498 TI - A spatially constrained mixture model for image segmentation VL - 16 T2 - Neural Networks, IEEE Transactions on SP - 494 KW - constrained KW - gaussian KW - gmm KW - mixture KW - segmentation KW - spatially AU - Blekas, K AU - Likas, A AU - Galatsanos, NP AU - Lagaris, IE PY - 2005/// UR - http://dx.doi.org/10.1109/TNN.2004.841773 ER -