In this work we describe the architecture of a functional system capable of processing Hartmann-Shack (H-S) images of the human eye and analyzing them automatically diagnose and measure the presence refractive errors in the eye under exam. Image analysis relies on connectionist Artificial Inteligence techniques such as Neural Nets, Fuzzy Logic and Classifier Combination. The major goal is to develop a new technology to measure ocular refrective errors (astigmatism, hypermetropia and myopia) that uses methods different from those adopted in patented systems. In general terms we are interested in investigaction approaches for solving current difficulties related to interpreting and diagnosing these and other ocular problems by analyzing obtained whit this acquisition technique.