Contourn descriptor generator algorithm implemented in embedded system

A contourn descriptor generator algorithm implemented in embedded system of new generation to obtain shape characterization of manufactured rigid objects is presented in the paper. Acquisition and processing stages to obtain information about the shape of rigid object by way of a descriptor vector are shown in order to be implemented in a single hardware piece processor embedded system to obtain parallel processing for multiple objects capabilities. To this purpose, the Boundary Object Function (BOF) is used and implemented in a piece of hardware of new generation to obtain a descriptor vector of any rigid manufacturing piece. The paper show results which makes the algorithm feasible to be used in applications of real time object manipulation for manufacturing flexible cells, the system uses a vision system with a reduced component implementation and low cost. In order to achieve the implementation of the algorithm a BeagleBone Black embedded system was choosen to demostrate the algorithm utility.