Evaluating Two Approaches for Farsi OCR Based on Sub-Word Shape Recognition
Subject Areas : electrical and computer engineering
1 -
2 - Tarbiat Modares University
Keywords: Farsi OCR sub-word shape clustering printed documents,
Abstract :
Two approaches for the recognition of printed Farsi documents based on sub-word shape recognition is proposed. First approach is based on recognition of sub-word shape as a whole and the second is based on the recognition of the body of sub-words. Sub-word body is constructed via removing dots and signs of the sub word. In second approach, information of dots and signs will be added after recognition of the body. Both approaches have two phases: training and test. In training phase, sub-words are clustered based on ISODATA algorithm. Initial centers of the clusters are computed through a hierarchical clustering algorithm. In first approach, sub-word recognition is performed in two stages: finding clusters close to the input sub-word and then finding the best match within the sub-words of these clusters. In the second approach another stage is required to find the final sub-word including dots and signs. Experimental results show that on clean images the first algorithm have better performance; 94% versus 93% in word level. But when dealing with low quality and noisy images, both algorithms are suffering from reduced accuracy. Sometimes this reduction is significant. The reasons of this behavior are inspected and some solutions are presented. Finally we compared both methods and inspected pros and cons of Farsi OCR based on sub-word shape.
[1] H. Y. Abdelazim and M. A. Hashish, "Arabic reading machine," in Proc. of the 10th National Computer Conf., Jeddah: Scientific Publishing Center, vol. 1, pp. 733-744, 1988.
[2] A. Amin, "Off-line Arabic character recognition: the state of the art," Pattern Recognition, vol. 31, no. 5, pp. 517-530, 1 Mar. 1998.
[3] M. Sarfraz, S. N. Nawaz, and A. Al - Khuraidly, "Offline Arabic text recognition system," in Proc. of Int. Conf. on Geometric Modeling and Graphics, pp. 30-35, 16-18 Jul. 2003.
[4] A. M. Gouda and M. A. Rashwan, "Segmentation of connected Arabic characters using hidden Markov models," in Prof. IEEE Int. Conf. on Computational Intelligence for Measurement Systems and Applications, pp. 115-119, 14-16 Jul. 2004.
[5] B. Kurdy and M. AlSabbagh, "Omnifont Arabic optical character recognition system," in Proc. of Int. Conf. on Information and Communication Technologies: from Theory to Applications, pp. 469-470, 19-23 Apr. 2004.
[6] A. I. Al - Shoshan, "Arabic OCR based on image invariants," in Proc. of the Int. Conf. on Geometric Modeling and Imaging- New Trends, pp. 150-154, London, UK, 16-18 Aug. 2006.
[7] س. ا. باقری یزدی و ب. ن. اعرابی، "استفاده همزمان از مدل مخفی مارکوف، برنامهنویسی پویا و ماشین بردار پشتیبان برای بازشناسی متن فارسی،" چهارمين كنفرانس بینایی ماشین و پردازش تصویر، مشهد، بهمن 1385.
[8] M. B. Menhaj and M. Adab, "Simultaneous segmentation and recognition of Farsi/Latin printed texts with MLP," in Proc. Int. Joint Conf. on Neural Networks, vol. 2, pp. 1534-1539, 12-17 May 2002.
[9] R. Mehran, H. Pirsiavash, and F. Razzazi, "A front-end OCR for omni-font Persian/Arabic cursive printed documents," Digital Imaging Computing: Techniques and Applications, p. 56, 6-08 Dec. 2005.
[10] R. Azmi and E. Kabir, "A new segmentation technique for omnifont Farsi text," Pattern Recognition Letters, vol. 22, no. 2, pp. 97-104, Feb. 2001.
[11] A. Ebrahimi and E. Kabir, "A pictorial dictionary for printed Farsi subwords," Pattern Recognition Letters, vol. 29, no. 5, pp. 656-663, Apr. 2008.
[12] B. Parhami and M. Taraghi, "Automatic recognition of printed Farsi texts," Pattern Recognition Letters, vol. 14, no. 1-6, pp. 395-403, 1981.
[13] ر. عزمی، بازشناسی متون چاپی فارسی، رساله دکتری، بخش مهندسی برق دانشگاه تربیت مدرس، 1378.
[14] س. رضوی و ا. کبیر، "یک پایگاه داده برای بازشناسی دستنوشتههای برخط فارسی،" ششمین كنفرانس سيستمهای هوشمند، صص. 225-218، كرمان، آذر 1383.
[15] H. Khosravi and E. Kabir, "Introducing a very large dataset of handwritten Farsi digits and a study on their varieties," Pattern Recognition Letters, vol. 28, no. 10, pp. 1133-1141, 15 Jul. 2007.
[16] G. H. Ball and D. I. Hall, "Some fundamental concepts and synthesis procedures for pattern recognition preprocessors," in Proc. Int. Conf. Microwaves, Circuit Theory, and Information Theory, pp. 113-114, Tokyo, Japan, 1964.
[17] ح. خسروی و ا. كبیر، "معرفی دو ویژگی سریع و كارآمد برای بازشناسی ارقام دستنویس فارسی،" چهارمين كنفرانس بینایی ماشین و پردازش تصویر، مشهد، صص 1131-1126، بهمن 1385.