Efficient Image Identifier Composition for Image Database

Abstract

As devices with image acquisition functionality have became affordable, the amount of images produced in diverse applications is enormous. Hence, binding an image with a distinctive value for the purpose identification needs to be efficient in the perspective of cost and effective regarding the goal as well. In this paper, we present a novel approach to image identifier generation. The proposed identifier generation method is motivated by pursuing a simple but effective and efficient approach. Taking fundamental image feature extraction methods into account, we make use of distribution of line segment so as to compose identifiers that basically satisfy one-to-one relationship between an image and a corresponding identifier. The generated identifiers can be used for name composition mechanism in a storage system or indexing in a massive image database. Our experimental results on generation of constituent index values have shown favorable results.