{"title":"Persian Printed Numerals Classification Using Extended Moment Invariants","authors":"Hamid Reza Boveiri","volume":39,"journal":"International Journal of Computer and Information Engineering","pagesStart":384,"pagesEnd":392,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/13920","abstract":"Classification of Persian printed numeral characters\r\nhas been considered and a proposed system has been introduced. In\r\nrepresentation stage, for the first time in Persian optical character\r\nrecognition, extended moment invariants has been utilized as\r\ncharacters image descriptor. In classification stage, four different\r\nclassifiers namely minimum mean distance, nearest neighbor rule,\r\nmulti layer perceptron, and fuzzy min-max neural network has been\r\nused, which first and second are traditional nonparametric statistical\r\nclassifier. Third is a well-known neural network and forth is a kind of\r\nfuzzy neural network that is based on utilizing hyperbox fuzzy sets.\r\nSet of different experiments has been done and variety of results has\r\nbeen presented. The results showed that extended moment invariants\r\nare qualified as features to classify Persian printed numeral\r\ncharacters.","references":"[1] S. Teodoridis and K. Koutroumbas, Pattern Recognition, 2nd Ed., CA:\r\nAcademic Press, 2003.\r\n[2] A. Darvish, E. Kabir, and H. Khosravi, \"Application of Shape Matching\r\nin Persian Handwritten Digits Recognition,\" Modares Technology and\r\nEng., vol. 22, pp. 37-47, Dec 2005.\r\n[3] H. Ketabdar, \"Persian Handwritten Digits Recognition with Structural\r\nApproach,\" in Proc. 6th Iranian Conf. on Electrical Engineering, Tehran,\r\n1998, pp. 4.19-4.24.\r\n[4] V. Johari and M. Razavi, \"Fuzzy Recognition of Persian Handwritten\r\nDigits,\" in Proc. 1st Iranian Conf. on Machine Vision and Image\r\nProcessing, Birjand, 2001, pp. 144-151.\r\n[5] H. Mozafari, F. Rahgozar, and Sh. Sharif, \"Online Recognition of\r\nPersian Handwritten Digits,\" in Proc. 2nd Computer Society of Iran\r\nAnnu. Conf., Tehran, 1996, pp. 196-204.\r\n[6] N. Sadati, E. Sheikhi, and B. Nazari, \"Utilizing Fuzzy Logic in\r\nRecognition of Persian Handwritten Digits,\" in Proc. 3rd Iranian Conf.\r\non Electrical Engineering, Tehran, 1995, pp. 241-248.\r\n[7] K. Faez, E. Khotanzad, and M. H. Shiralishahreza, \"Recognition of\r\nPersian Handwritten Letters and Digits Using Pseudo-Zernike Moments\r\nand Neural Networks,\" in Proc. 3rd Iranian Conf. on Electrical\r\nEngineering, Tehran, 1995, pp. 231-240.\r\n[8] H. Nabavi, H. Sadoughi, E. Kabir, and Karlos, \"Improvement of Persian\r\nHandwritten Digits Recognition Using Composition of Classifiers with\r\nANFIS,\" in Proc. 5th Iranian Conf. on Fuzzy Systems, Tehran, 2004, pp.\r\n337-345.\r\n[9] K. Masrori and V. Pourmohseni, \"Recognition of Persian Handwritten\r\nDigits Using DWT Algorithm,\" in Proc. 4th Nat. Computer Society of\r\nIran Annu. Conf., Tehran, 1998, pp. 1-7.\r\n[10] L. C. Liu and C. Y. Suen, \"A new benchmark on the recognition of\r\nhandwritten Bangla and Farsi numeral characters,\" Pattern Recognition,\r\nto be published.\r\n[11] A. Harifi and A. Aghagolzadeh, \"A New Pattern for Handwritten\r\nPersian\/Arabic digit Recognition,\" in Proc. of World Academy of Sci.,\r\nEng. and Technology, vol. 3, pp. 130-133, Jan. 2005.\r\n[12] S. Mozaffari, K. Faez, and M. Ziaratban, \"A Hybrid Structural\/Statistical\r\nClassifier for Handwritten Farsi\/Arabic Numeral Recognition,\" in IAPR\r\nConf. on Machine Vision Application, Tsukuba, 2005, pp. 218-221.\r\n[13] A. Mowlaei and K. Faez, \"Recognition of Isolated Handwritten\r\nPersian\/Arabic Characters and Numerals Using Support Vector\r\nMachine,\" IEEE XIII Workshop on Neural Networks for Signal\r\nProcessing, 2003, pp. 547-554.\r\n[14] J. Sadri, C. Y. Suen, and T. D. Bui, \"Application of Support Vector\r\nMachines for Recognition of Handwritten Arabic\/Persian Digits,\" in\r\nProc. 2nd Conf. on Machine Vision and Image Processing, Tehran, 2003,\r\npp. 300-307.\r\n[15] H. Khosravi and E. Kabir, \"Introducing a very large dataset of\r\nhandwritten Farsi digits and a study on their varieties,\" Pattern\r\nRecognition Lett., vol. 28, pp. 1133-1141, Dec. 2007.\r\n[16] F. Soleimanipour, J. Sadri, and C. Y. Suen, \"Standard Databases for\r\nRecognition of Handwritten Digits, Numerical Strings, Legal Amounts,\r\nLetters and Dates in Farsi Language,\" in Proc. 10th Int. Workshop on\r\nFrontiers of Handwritten Recognition, La baule, 2006, pp. 3-7.\r\n[17] S. Mozaffari, K. Faez, F. Faradji, M. Ziaratban, and S. M. Golzan, \"A\r\nComprehensive Isolated Farsi\/Arabic Character Database for\r\nHandwritten OCR Research,\" in Proc. 10th Int. Workshop on Frontiers\r\nof Handwritten Recognition, La baule, 2006, pp. 379-384.\r\n[18] H. Ebrahimnezhad and Gh. Montazer, \"Presenting Improved Fuzzy\r\nApproach for Persian Numeral Fonts Using Entropy Function,\" in Proc.\r\n11th Nat. Computer Society of Iran Conf., Tehran, 2006, pp. 883-889.\r\n[19] H. Ebrahimnezhad, Gh. A. Montazer, and N. Jafari, \"Recognition of\r\nPersian Numeral Fonts by Combining the Entropy Minimized Fuzzifier\r\nand Fuzzy Grammar,\" in Proc. 6th PWASET Int. Conf. on Artificial\r\nIntelligence, Knowledge Engineering and Data Bases, Corfu Island,\r\n2007, pp. 22-27.\r\n[20] H. R. Boveiri, \"Persian Printed Numeral Characters Recognition Using\r\nGeometrical Central Moments and Fuzzy Min-Max Neural Network,\"\r\nIEEE Trans. Neural Netw., submitted for publication.\r\n[21] H. R. Boveiri, \"Classification of Persian Printed digits Using Moment\r\nInvariants and Fuzzy Min-Max Neural Network,\" 14th International CSI\r\nConf., Tehran, 2009, submitted for publication.\r\n[22] M. K. Hu, \"Visual Pattern Recognition by Moment Invariant,\" IRE\r\nTrans. Info. Theory, vol. IT - 8, pp. 179-187, Feb 1962.\r\n[23] Y. Li, \"Reforming the Theory of Invariant Moments for Pattern\r\nRecognition,\" Pattern Recognition Lett., vol. 25, pp. 723-730, Jul 1992.\r\n[24] H. R. Boveiri, \"Scanned Persian Printed Text Characters Recognition\r\nUsing Fuzzy-Neural Networks,\" M. S. thesis, Islamic Azad Univ.\r\nScience and Research Branch, Ahvaz, 2009.\r\n[25] S. Alirezaee, H. Aghaeinia, M. Ahmadi, and K. Faez, \"Recognition of\r\nMiddle Age Persian characters Using a Set of Invariant Moments,\" in\r\nProc. 33rd Applied Imagery Pattern Recognition Workshop, Washington\r\nDC, 2004, pp. 196-201.\r\n[26] B. Attaran and M. Jahed, \"Extraction and Recognition of letters and\r\ndigits in Vehicle License Plate Using Sobel Filter, Neural Networks, and\r\nHu Moments,\" in Proc. 1st Iranian Conf. on Machine Vision and Image\r\nProcessing, Birjand, 2001, pp. 287-299.\r\n[27] A. Khotanzad and J-H. Lu, \"Classification of Invariant Image\r\nRepresentations using a Neural Networks,\" IEEE Trans. on Acoust.,\r\nSpeech and Signal Process., vol. 38, no. 6, pp. 1028-1038, Jun. 1990.\r\n[28] C. R. Gonzalez and E. R. Woods, Digital Image Processing, 2nd Ed.,\r\nNew Jersey: Prentice-Hall Inc., 2002.\r\n[29] M. M. Gupta, L. Jin, and N. Homma, Static and Dynamic Neural\r\nNetworks: From Fundamentals to Advance theory, New Jersey: John\r\nWiley & Sons Inc., 2003.\r\n[30] P. K. Simpson, \"Fuzzy Min-Max Neural Networks- Part1:\r\nClassification,\" IEEE Trans. Neural Netw., vol. 3, no. 5, pp. 776-786,\r\nSep. 1992.\r\n[31] P. K. Simpson, \"Fuzzy Min-Max Neural Networks- Part2: Clustering,\"\r\nIEEE Trans. Fuzzy Syst., vol. 1, no. 1, pp. 32-45, Feb. 1993.","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 39, 2010"}