Browsing Heterogeneous Document Collections by a Segmentation-Free Word Spotting Method
Authors:
- Marcal RusinolMRMarcal Rusinol
- David AldavertDADavid Aldavert
- Ricardo ToledoRTRicardo Toledo
- Josep LladosJLJosep Llados
- 1 September 2011
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15205363,p. 63-67
Abstract
In this paper, we present a segmentation-free word spotting method that is able to deal with heterogeneous document image collections. We propose a patch-based framework where patches are represented by a bag-of-visual-words model powered by SIFT descriptors. A later refinement of the feature vectors is performed by applying the latent semantic indexing technique. The proposed method performs well on both handwritten and typewritten historical document images. We have also tested our method on documents written in non-Latin scripts.Keywords
This publication has 10 references indexed in Scilit:
- HMM-based Word Spotting in Handwritten Documents Using Subword Models2010
- A Novel Word Spotting Algorithm Using Bidirectional Long Short-Term Memory Neural Networks2010
- Towards an omnilingual word retrieval system for ancient manuscriptsPattern Recognition, 2009
- Segmentation-free Word Spotting in Historical Printed Documents2009
- Keyword Spotting in Document Images through Word Shape Coding2009
- Local Features and Kernels for Classification of Texture and Object Categories: A Comprehensive StudyInternational Journal of Computer Vision, 2006
- Word spotting for historical documentsInternational Journal on Document Analysis and Recognition (IJDAR), 2006
- Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories2006
- Content-Based Retrieval of Historical Ottoman Documents Stored as Textual ImagesIEEE Transactions on Image Processing, 2004
- Indexing by latent semantic analysisJournal of the American Society for Information Science, 1990