Content Based Image Retrieval Using Adaptive Inverse Pyramid Representation

摘要

This paper presents a new approach for content-based image retrieval using cognitive representation with pyramidal decomposition. This approach corresponds to the hypothesis of the human way for object recognition based on consecutive approximations with increased resolution for the selected regions of interest. The method is based on object model creation with Inverse Difference Pyramid controlled by neural network. The method's basic advantages are the high flexibility and the ability to create general models for various views and scaling with relatively low computational complexity. The method is suitable for great number of applications - medicine, digital libraries, electronic galleries, geographic information systems, documents archiving, digital communication systems, etc.