Computer simulation for optical pattern recognition of three-dimensional object perspective invariant

By using a bank of filters in the conventional optical correlation system instead of a single filter, the three-dimensional object recognition based on the optical pattern recognition of 3-D object perspective invariant was realized with improved recognition precision by computer simulation. The difficulties in the design of filters bank were overcome by extracting the eigen-information from targets of different perspectives and by deriving the design equations as well as the restrained conditions of the filters bank of three object perspective invariant.