DIRECT SHAPE FROM SHADING WITH IMPROVED RATE OF CONVERGENCE

In this paper, new direct shape from shading algorithms with improved rate of convergence for inclined and perpendicular light source are pr esented. The algorithms are mainly based on optimal dynamic system con trol with dynamic programming. Unlike the previous shading algorithms which first estimate the gradient of surface instead of depth using va riational approaches, our algorithms provide a direct relative depth p erception. Here the maximum uphill principle for inclined and perpendi cular light source is derived instead of using the existing minimum do wnhill criteria in order to perceive the normal shape of the object. A lso the convergence conditions for the algorithms are derived and prov en under zero-initial condition. We show that the learning rate is lin early proportional to the minimum value of the input image in the disc rete case. Furthermore, in order to increase the rate of convergence, the momentum factor is introduced. (C) 1997 Pattern Recognition Societ y.