Most of the existing methods for optical how estimation are based on a
constraint equation which is defined for each image pixel. This class
of algorithms is usually called gradient-based. Due to the structure
of the constraint equation, the problem is ill-posed, thus some soluti
ons based on regularization have been proposed in the past. On the con
trary, if other constraint equations can be found for the pixel under
consideration or in its immediate neighborhood, the problem is not ill
-posed and a solution can be found by solving determined or over-deter
mined systems of equations. Following this reasoning, several algorith
ms for evaluating the optical how have been proposed in the literature
. Most of these over-determined systems of equations are solved by usi
ng least-squares techniques. In this paper, a new approach is presente
d in order to eliminate, or strongly reduce, the drawbacks of least-sq
uares and regularization-based techniques. This is based on a modified
version of the Combinatorial Hough Transform. A comparison is made be
tween the results obtained with the new approach and those produced by
the classical least-squares and regularization-based techniques. (C)
1995 Academic Press, Inc.