Three-dimensional object recognition based on multiple perspectives imaging with microlens arrays

We describe a novel optical system to recognize 3D objects in real time by using multiple perspectives based on integral photography. Multiple perspectives of the 3D objects are obtained by using microlens arrays. Then, an optical processor in cascade performs the correlation between the 3D objects in real time. The proposed system, consists of two subsystems. One is a projection system of a 3D object into an array of 2D images that contain different perspectives of the 3D object and the other is an optical processor to perform the 2D correlation. In the projection system, similar to those used in integral photography, a 3D object under incoherent illumination is projected as an array of 2D elemental images by a periodic microlens array.