Two-dimensional vector modeling of image random field using artificial neural networks

The problem of 2-D vector modeling of an image random field using a neural network approach is addressed. A new learning scheme is developed using the recursive least squares (RLS) method which can be employed to extract the vector model coefficients. 2-D vector autoregressive models with various causal and noncausal regions of support (ROS) are considered. The proposed scheme is inherently fast and ideally suited for real-time implementations. It does not need any prior statistical knowledge of the image process or any matrix manipulation. Numerical results demonstrate the advantages of the proposed scheme over the conventional parameter estimation methods