The aim of this paper is to analyse the possibilities of applying signal processing and artificial vision techniques to the ore sorting problem. In order to overcome the limitations imposed by the classical approach to the problem, which essentialy consists of using on}y punctual information, neglecting the relationships among diffe- rent points, we are studying these other methods.
This paper also describes the preliminary works of design and the co- rresponding experimentation of an artificial vision system, dealing with coal sorting problems.