Bøger i Research Notes in Neural Computing serien

In this book a global shape model is developed and applied to the analysis of real pictures acquired with a visible light camera under varying conditions of optical degradation.

These functions range from such basic ones as detection of changes that may lead to tissue damage and eventual destruction of the organism and the implementation of avoidance reactions, to more elaborate representations of the external world implying recognition of shapes, sounds and textures as the basis of planned action or even reflection.

This volume integrates theory and experiment to place the study of vision within the context of the action systems which use visual information.

The study of neural networks is enjoying a great renaissance, both in computational neuroscience - the development of information processing models of living brains - and in neural computing - the use of neurally inspired concepts in the construction of "intelligent" machines.

This monograph is an outgrowth of the authors' recent research on the de- velopment of algorithms for several low-level vision problems using artificial neural networks. Specific problems considered are static and motion stereo, computation of optical flow, and deblurring an image. From a mathematical point of view, these inverse problems are ill-posed according to Hadamard. Researchers in computer vision have taken the "e;regularization"e; approach to these problems, where one comes up with an appropriate energy or cost function and finds a minimum. Additional constraints such as smoothness, integrability of surfaces, and preservation of discontinuities are added to the cost function explicitly or implicitly. Depending on the nature of the inver- sion to be performed and the constraints, the cost function could exhibit several minima. Optimization of such nonconvex functions can be quite involved. Although progress has been made in making techniques such as simulated annealing computationally more reasonable, it is our view that one can often find satisfactory solutions using deterministic optimization algorithms.