K-D Indexing in Printed Trilingual Documen

Abstract

Abstract-In this work, we proposed indexing model for script identification. Rectangular White Space analysis algorithm is used to analyze and identify heterogeneous layouts of document images. To speed up the script identification, we focus on designing an indexing mechanism for tri-lingual scripts for optimizing the subsequent robust identification system. For representation, we extract features from Gabor responses and also using scale invariant feature transform. We considered a set of global features and index by Kd-tree. For experimentation, we have used our own database. Experimental results reveal that indexing prior to identification is faster than conventional identification method in terms of time for scripts.