An oscillatory elastic graph matching model for scene analysis

摘要

Scene analysis so far is one of the most important topics in machine vision. In this paper, we present an integrated Neurla Oscillatory Model together with an Elastic Graph Dynamci Link Model to provide an automatic scene analysis system for color images. The model involves: 1) Multi-frequency bands feature extraction scheme using Gabor Filters, 2) Automatic figures-ground object segmentaiton using composite neural oscillatory model, 3) Object matching using Elastic Graph Dynamci Link Model. Using an image gallery of over 3000 color objects, with the recognitin of 6000 different scenes, our model shows an average correct recognition rate of over 95