Isolated Persian/Arabic word spotting by label embedding

  • majid Iranpour Mobarakeh Shahrood university
  • Alireza Ahmadyfard Shahrood university
Keywords: Word Retrieval, Word Spotting, Attribute-based Classification, Label Embedding,

Abstract

The goal of word image retrieval and word spotting, as a particular case of semantic content based image retrieval (CBIR) is to find image or text(string) query word in a dataset of images. In this paper we evaluate a holistic approach for Persian handwritten isolated word image retrieval that can be used for word recognition. The aim of recognition is to recognize the content of word image, usually aided by a dictionary or lexicon. To the best of our knowledge, this paper is the first paper about Persian script. This is achieved by a combination of label embedding, attribute based classification and common subspace regression. In this subspace, image and string representation of the same word are close together that allows one to cast recognition and retrieval task as a nearest neighbor problem. Unlike other existing methods in word image retrieval and word spotting, the representation has a fixed length, low dimensionality and it is very fast to compute and compare. We used Farsa and Iranshahr, two common datasets of isolated Persian handwritten words to evaluate method. The Experiment shows promising results for word image retrieval and word spotting and respectively achieved 100% and 97% of recognition rate for Farsa and Iranshahr datasets.

References

David, J. E. Y. W. T.& Maire, F. R. B. M.&Vesom, G. 2005. Making latin manuscripts searchable using gHMM’s. Advances in Neural Information Processing Systems 17: Proceedings of the 2004 Conference,

Lu, S.&Tan, C. L. 2008. Retrieval of machine-printed Latin documents through Word Shape Coding. Pattern Recognition. Pp 1799-1809

Srihari, S. N.& Srinivasan, H.& Huang, C.&Shetty, S. 2006. Spotting words in Latin, Devanagari and Arabic scripts. VIVEK-BOMBAY-. Pp 2

Huang, L.& Yin, F.& Chen, Q.-H.&Liu, C.-L. 2013. Keyword spotting in unconstrained handwritten Chinese documents using contextual word model. Image and Vision Computing. Pp 958-968

Lu, Y.&Tan, C. L. 2002. Word spotting in Chinese document images without layout analysis. Pattern Recognition, 2002. Proceedings. 16th International Conference on,

Lu, Y.&Tan, C. L. 2004. Chinese word searching in imaged documents. International Journal of Pattern Recognition and Artificial Intelligence. Pp 229-246

Aouadi, N.&Kacem, A. 2011. Word spotting for Arabic handwritten historical document retrieval using generalized hough transform. PATTERNS 2011, The Third International Conferences on Pervasive Patterns and Applications,

Kefali, A.&Chemmam, C. 2011. A Semi-Automatic Approach of old Arabic Documents Indexing. CIIA,

Khayyat, M.& Lam, L.&Suen, C. Y. 2014. Learning-based word spotting system for Arabic handwritten documents. Pattern Recognition. Pp 1021-1030

Pourasad, Y.& Hassibi, H.&Ghorbani, A. 2013. A word spotting method for Farsi machine-printed document images. Turkish Journal of Electrical Engineering & Computer Sciences. Pp 734-746

POURASAD, Y.& HASSIBI, H.&GHORBANI, A. 2012. A Farsi/Arabic Word Spotting Approach for Printed Document Images. International Journal of Natural & Engineering Sciences. Pp

Azmi, R. 2011. Farsi/Arabic Document Image Retrieval through Sub-Letter Shape Coding for mixed Farsi/Arabic and English text. International Journal of Computer Science Issues (IJCSI). Pp

Almazán, J.& Gordo, A.& Fornés, A.&Valveny, E. 2013. Handwritten Word Spotting with Corrected Attributes. ICCV 2013-IEEE International Conference on Computer Vision (2013),

Lodhi, H.& Saunders, C.& Shawe-Taylor, J.& Cristianini, N.&Watkins, C. 2002. Text classification using string kernels. The Journal of Machine Learning Research. Pp 419-444

Leslie, C. S.& Eskin, E.&Noble, W. S. 2002. The spectrum kernel: A string kernel for SVM protein classification. Pacific symposium on biocomputing,

Almazán, J.& Gordo, A.& Fornés, A.&Valveny, E. 2014a. Word Spotting and Recognition with Embedded Attributes. Pp

Jegou, H.& Douze, M.&Schmid, C. 2011. Product quantization for nearest neighbor search. Pattern Analysis and Machine Intelligence, IEEE Transactions on. Pp 117-128

Fischer, A.& Keller, A.& Frinken, V.&Bunke, H. 2010. HMM-based word spotting in handwritten documents using subword models. Pattern recognition (icpr), 2010 20th international conference on,

Frinken, V.& Fischer, A.& Manmatha, R.&Bunke, H. 2012. A novel word spotting method based on recurrent neural networks. Pattern Analysis and Machine Intelligence, IEEE Transactions on. Pp 211-224

Manmatha, R.& Han, C.&Riseman, E. M. 1996. Word spotting: A new approach to indexing handwriting. Computer Vision and Pattern Recognition, 1996. Proceedings CVPR'96, 1996 IEEE Computer Society Conference on,

Rath, T. M.& Manmatha, R.&Lavrenko, V. 2004. A search engine for historical manuscript images. Proceedings of the 27th annual international ACM SIGIR conference on Research and development in information retrieval,

Rath, T. M.&Manmatha, R. 2007. Word spotting for historical documents. International Journal of Document Analysis and Recognition (IJDAR). Pp 139-152

Rodrıguez, J. A.&Perronnin, F. 2008. Local gradient histogram features for word spotting in unconstrained handwritten documents. Int. Conf. on Frontiers in Handwriting Recognition,

Rodríguez-Serrano, J. A.&Perronnin, F. 2009. Handwritten word-spotting using hidden Markov models and universal vocabularies. Pattern Recognition. Pp 2106-2116

Rodríguez-Serrano, J. A.&Perronnin, F. 2012. A model-based sequence similarity with application to handwritten word spotting. Pattern Analysis and Machine Intelligence, IEEE Transactions on. Pp 2108-2120

Yalniz, I. Z.&Manmatha, R. 2011. An efficient framework for searching text in noisy document images. Pp

Keaton, P.& Greenspan, H.&Goodman, R. 1997. Keyword spotting for cursive document retrieval. Document Image Analysis, 1997.(DIA'97) Proceedings., Workshop on,

Perronnin, F.&Rodriguez-Serrano, J. A. 2009. Fisher kernels for handwritten word-spotting. Document Analysis and Recognition, 2009. ICDAR'09. 10th International Conference on,

Gatos, B.&Pratikakis, I. 2009. Segmentation-free word spotting in historical printed documents. Document Analysis and Recognition, 2009. ICDAR'09. 10th International Conference on,

Rusinol, M.& Aldavert, D.& Toledo, R.&Lladós, J. 2011. Browsing heterogeneous document collections by a segmentation-free word spotting method. Document Analysis and Recognition (ICDAR), 2011 International Conference on,

Almazán, J.& Gordo, A.& Fornés, A.&Valveny, E. 2012. Efficient exemplar word spotting. Pp

Almazán, J.& Gordo, A.& Fornés, A.&Valveny, E. 2014b. Segmentation-free word spotting with exemplar SVMs. Pattern Recognition. Pp

Dalal, N.&Triggs, B. 2005. Histograms of oriented gradients for human detection. Computer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer Society Conference on,

Malisiewicz, T.& Gupta, A.&Efros, A. A. 2011. Ensemble of exemplar-svms for object detection and beyond. Computer Vision (ICCV), 2011 IEEE International Conference on,

Fischer, A.& Keller, A.& Frinken, V.&Bunke, H. 2012. Lexicon-free handwritten word spotting using character HMMs. Pattern Recognition Letters. Pp 934-942

Moghaddam, R. F.&Cheriet, M. 2009. Application of multi-level classifiers and clustering for automatic word spotting in historical document images. Document Analysis and Recognition, 2009. ICDAR'09. 10th International Conference on,

Saabni, R.&Bronstein, A. 2011. Fast key-word searching via embedding and Active-DTW. Document Analysis and Recognition (ICDAR), 2011 International Conference on,

Sridha, M.& Mandalapu, D.&Patel, M. 2006. Active-dtw: a generative classifier that combines elastic matching with active shape modeling for online handwritten character recognition. Tenth International Workshop on Frontiers in Handwriting Recognition,

Sari, T.&Kefali, A. 2008. A search engine for Arabic documents. Dixième Colloque International Francophone sur l'Ecrit et le Document,

Saykol, E.& Sinop, A. K.& Gudukbay, U.& Ulusoy, O.&Çetin, A. E. 2004. Content-based retrieval of historical Ottoman documents stored as textual images. Image Processing, IEEE Transactions on. Pp 314-325

Shahab, S.& Al-Khatib, W. G.&Mahmoud, S. A. 2006. Computer aided indexing of historical manuscripts. Computer Graphics, Imaging and Visualisation, 2006 International Conference on,

Farhadi, A.& Endres, I.& Hoiem, D.&Forsyth, D. 2009. Describing objects by their attributes. Computer Vision and Pattern Recognition, 2009. CVPR 2009. IEEE Conference on,

Akata, Z.& Perronnin, F.& Harchaoui, Z.&Schmid, C. 2013a. Label-embedding for attribute-based classification. Computer Vision and Pattern Recognition (CVPR), 2013 IEEE Conference on,

Akata, Z.& Perronnin, F.& Harchaoui, Z.&Schmid, C. 2013b. Attribute-Based Classification with Label-Embedding. NIPS 2013 Workshop on Output Representation Learning,

Ferrari, V.&Zisserman, A. 2008. Learning visual attributes. Pp

Lampert, C.& Nickisch, H.&Harmeling, S. 2013. Attribute-based classification for zero-shot learning of object categories. Pp

Scheirer, W. J.& Kumar, N.& Belhumeur, P. N.&Boult, T. E. 2012. Multi-attribute spaces: Calibration for attribute fusion and similarity search. Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on,

Hardoon, D. R.& Szedmak, S.&Shawe-Taylor, J. 2004. Canonical correlation analysis: An overview with application to learning methods. Neural computation. Pp 2639-2664

Blaschko, M. B.&Lampert, C. H. 2008. Correlational spectral clustering. Computer Vision and Pattern Recognition, 2008. CVPR 2008. IEEE Conference on,

Gong, Y.& Lazebnik, S.& Gordo, A.&Perronnin, F. 2013. Iterative quantization: A procrustean approach to learning binary codes for large-scale image retrieval. Pattern Analysis and Machine Intelligence, IEEE Transactions on. Pp 2916-2929

Rodriguez-Serrano, J. A.& Perronnin, F.&Meylan, F. Label embedding for text recognition. Pp

Z.Imani& A.R.Ahmadifard&A.Zohrevand. 2013. Introduction to Database FARSA: digital images of handwritten Farsi words. Eleventh Iranian Conference on Intelligent Systems (ICIS2013), Bam, Iran.

Dehghan, M.& Faez, K.& Ahmadi, M.&Shridhar, M. 2001a. Handwritten Farsi (Arabic) word recognition: a holistic approach using discrete HMM. Pattern Recognition. Pp 1057-1065

Dehghan, M.& Faez, K.& Ahmadi, M.&Shridhar, M. 2001b. Unconstrained Farsi handwritten word recognition using fuzzy vector quantization and hidden Markov models. Pattern Recognition Letters. Pp 209-214

Imani, Z.& Ahmadyfard, A.& Zohrevand, A.&Alipour, M. 2013. Offline handwritten Farsi cursive text recognition using hidden Markov models. Machine Vision and Image Processing (MVIP), 2013 8th Iranian Conference on,

Ebrahimpour Reza, S. S., Javadi Mehrdad, Sharifizadeh Fatemeh. 2011. Handwritten Word Recognition by Multiple Classifiers: A Divide-and-Conquer Approach. International Journal of Applied Research on Information Technology and Computing. Pp 9-21

Ebrahimpour, R.& Sarhangi, S.&Sharifizadeh, F. 2011. Mixture of Experts for Persian handwritten word recognition. Iranian Journal of Electrical & Electronic Engineering. Pp 217-224

Published
2017-01-16
Section
Electrical Engineering (1)