A spatially constrained mixture model for image segmentation

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.

[1]  Thomas J. Hebert,et al.  Bayesian pixel classification using spatially variant finite mixtures and the generalized EM algorithm , 1998, IEEE Trans. Image Process..

[2]  Stephen M. Smith,et al.  Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm , 2001, IEEE Transactions on Medical Imaging.

[3]  Pierre Baldi,et al.  Improving the prediction of protein secondary structure in three and eight classes using recurrent neural networks and profiles , 2002, Proteins.

[4]  Volker Tresp,et al.  Averaging, maximum penalized likelihood and Bayesian estimation for improving Gaussian mixture probability density estimates , 1998, IEEE Trans. Neural Networks.

[5]  Peter Tiño,et al.  Recurrent Neural Networks with Small Weights Implement Definite Memory Machines , 2003, Neural Computation.

[6]  Peter Tiño,et al.  Architectural Bias in Recurrent Neural Networks: Fractal Analysis , 2002, Neural Computation.

[7]  J. Kolen Recurrent Networks: State Machines Or Iterated Function Systems? , 1994 .

[8]  Jürgen Schmidhuber,et al.  Learning Nonregular Languages: A Comparison of Simple Recurrent Networks and LSTM , 2002, Neural Computation.

[9]  Stephen J. Wright,et al.  Numerical Optimization , 2018, Fundamental Statistical Inference.

[10]  Peter Ti Markovian Architectural Bias of Recurrent Neural Networks , 2004 .

[11]  Nuggehally Sampath Jayant,et al.  An adaptive clustering algorithm for image segmentation , 1989, International Conference on Acoustics, Speech, and Signal Processing,.

[12]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[13]  P. Green Bayesian reconstructions from emission tomography data using a modified EM algorithm. , 1990, IEEE transactions on medical imaging.

[14]  Sankar K. Pal,et al.  A review on image segmentation techniques , 1993, Pattern Recognit..

[15]  DeLiang Wang,et al.  Weight adaptation and oscillatory correlation for image segmentation , 2000, IEEE Trans. Neural Networks Learn. Syst..