Composite filters based on digital holograms for distortion-tolerant three-dimensional object recognition

The use of optoelectronic systems for information processing has been widely explored. In particular, a lot of work has been done on 2D pattern recognition using optical correlation techniques. Recently, these techniques have been extended to be used in conjunction with digital holography in order to recognize 3D objects. In this presentation, we show how it is possible to create nonlinear composite filters with various views of a 3D reference object reconstructed from digital holograms. These composite filters allow us to achieve distortion-tolerant recognition. As an example, we demonstrate tolerance against out-of-plane rotation and longitudinal shift along the z-axis.