Farsi/Arabic Document Image Retrieval Through Sub-Letter Shape Coding

Abstract

A retrieval method for explicit recognition free Farsi/Arabic document is proposed in this paper. The system can be used in mixed Farsi/Arabic and English text. The method consists of Preprocessing, word and sub_word extraction, detection and cancelation of sub_letter connectors, annotation sub_letters by shape coding, classifier of sub_letters by use of decision tree and using of RBF neural network for sub_letter recognition. The Proposed system retrieves document images by a new sub_letter shape coding scheme in Farsi/Arabic documents. In this method document content captures through sub_letter coding of words. The decision tree-based classifier partitions the sub_letters space into a number of sub regions by splitting the sub_letter space, using one topological shape features at a time. Topological shape Features include height, width, holes, openings, valleys, jags, sub_letter ascenders/descanters. Experimental results show advantages of this method in Farsi/Arabic Document Image Retrieval.