Indexing for Mongolian Kanjur images in word spotting

Hongxi Wei,Guanglai Gao,Xueliang Zhang
2013-01-01
Journal of Computational Information Systems
Abstract:Retrieving historical document images is still a challenging task. In this paper, a method for indexing word images of the Mongolian Kanjur has been presented under the framework of word spotting. In order to accomplish the real-time image-to-image matching at the retrieval stage, the Discrete Fourier Transform is performed on each profile feature of word images. An appropriate number of moduli of the lower-order DFT complex coefficients can be obtained. In this way, each word image of the Mongolian Kanjur can be converted into a fixed-length feature vector by combining the moduli of each profile feature, which is taken as an indexing term. Thus, the procedure of locating word images can be realized by means of calculating similarities between the fixed-length feature vectors of a given query word image and each word image of the Mongolian Kanjur. Copyright © 2013 Binary Information Press.
What problem does this paper attempt to address?