A pattern recognition system based on models of aspects of thehuman visual system

摘要

The natural world contains many examples of successful visionsystems embodying important design principles and remarkableoptimalities. There is much to be learned by studying and modellingnatural vision systems. The algorithm discussed illustrates how aspectsof even the present partial understanding of human vision can be appliedto pattern representation and recognition. The authors review thelog-polar model for the sampling structure of the human retina, thenconsider the application of the model in an algorithm for scale androtation invariant recognition of patterns