In this paper, we present a probabilistic neural model, which extends Kohonen's self-organizing map (SOM) by performing a probabilistic principal component analysis (PPCA) at each neuron. Several SOMs have been proposed in the literature to capture the local principal subspaces, but our approach offers a probabilistic model while it has a low complexity on the dimensionality of the input space. This allows to process very high-dimensional data to obtain reliable estimations of the probability densities which are based on the PPCA framework. Experimental results are presented, which show the map formation capabilities of the proposal with high-dimensional data, and its potential in image and video compression applications.

No abstract

No abstract

No abstract

No abstract

The effective processing of visual data without interruption is currently of supreme importance. For that purpose, the analysis system must adapt to events that may affect the data quality and maintain its performance level over time. A methodology for background modeling and foreground detection, whose main characteristic is its robustness against stationary noise, is presented in the paper. The system is based on a stacked denoising autoencoder which extracts a set of significant features for each patch of several shifted tilings of the video frame. A probabilistic model for each patch is learned. The distinct patches which include a particular pixel are considered for that pixel classification. The experiments show that classical methods existing in the literature experience drastic performance drops when noise is present in the video sequences, whereas the proposed one seems to be slightly affected. This fact corroborates the idea of robustness of our proposal, in addition to its usefulness for the processing and analysis of continuous data during uninterrupted periods of time.

We present a new neural model which extends the classical competitive learning (CL) by performing a Probabilistic Principal Components Analysis (PPCA) at each neuron. The model also has the ability to learn the number of basis vectors required to represent the principal directions of each cluster, so it overcomes a drawback of most local PCA models, where the dimensionality of a cluster must be fixed a priori. Experimental results are presented to show the performance of the network with multispectral image data.

No abstract