Individuality Representation in Character Recognition
DOI:
https://doi.org/10.21928/juhd.v1n2y2015.pp300-305Abstract
The task of recognition that is based on handwriting characters in the Kurdish language is an interesting study in the area of computer vision and pattern recognition. In the past couple of years, numerous state-of-the-art techniques and methods have been created for pattern recognition. On the other hand, Kurdish language handwriting recognition has been seen to be more difficult when compared to other different languages. The similarities in the properties in Kurdish characters is the primary reason of the great resemblance in the features of Kurdish handwriting characters, therefore the requirement for the recognition process is critical. Consequently, to obtain accurate and precise recognition on the basis of the Kurdish handwriting character, it is crucial for the resemblances in the character properties of Kurdish handwriting to be distinguished. To identify a particular character, the style of character handwriting may be evaluated to enable the implied representation of the hidden unique features of the user’s character. Unique features may guide in recognizing characters that may be important when recognizing the correct character among similar characters. On the other hand, the problem of the resemblances in the properties of handwriting of Kurdish characters were not taken into account ,consequently leaving a high chance of reducing the similarity error for any intra-class (of the same character),with the reduction of the similarity error for any inter-class (of different characters) as well. In order to obtain higher effectiveness, this study uses discretization features for reducing the similarity error for intra-class (of the same character),with the increase of the similarity error for inter-class (of different characters)in recognition of Kurdish Handwriting characters with MAE.
References
[2] Muzaffar, B. and Jurgen, K.,(2009). Person Authentication with RDTW based on Handwritten PIN and Signature with a Novel Biometric Smart Pen Device, IEEE
Workshop. 63-68.
[3] Srihari, N. S. and Ball, R. G., (2009). Semi-Supervised Learning for Handwriting
Recognition. ICDAR, 26-30.
[4] Tonghua, S., Zhang, T-W., Guan, D.J., Huang, H.J.,(2009). Off-Line Recognition of Realistic Chinese Handwriting using Segmentation-Free Strategy. Pattern Recognition. 42(1), 167-182.
[5] Behzad, H. and Mohsen, M.,(2010). A text-independent Persian Writer Identification based on Feature Relation Graph (FRG). Pattern Recognition 43. 2199–2209.
[6] Bayan Omar Mohammed , (2013). HANDWRITTEN KURDISH CHARACTER RECOGNITION USING GEOMETRIC DISCERTIZATION FEATURE , Volume 4, Number 1 , January-June 2013 pp. 51-55.
[7] Anil K. J., Robert P. W. and Jianchang, D. M. (2000). Statistical Pattern Recognition: A Review, in Proc. 4th IEEE Trans on Pattern analysis and Machine intelligence, 22, 4-37.
[8] AK Muda, SM Shamsuddin, A Ajith,(2010). Improvement of Authorship invarianceness for individuality representation in writer identification. Neural Netw. World 3(10), 371–387.
[9] Bayan Omar Mohammed , (2012). ‘ Uniqueness in Kurdish Handwriting’, International Journal of Engineering & Computer Science IJECS-IJENS Vol:12 No:06 , pp:42-50.
[10] Fabrice Muhlenbach and Ricco Rakotomalala, (2005). Discretization of Continuous Attributes. In John Wang (Ed.) Encyclopedi,a of Data Warehousing and
Mining, pp. 397-402.
[11] B. O. Mohammed , S. M. Shamsuddin ,(2012). Improvement in twins handwriting identification with invariants discretization , EURASIP Journal on Advances in Signal Processing 2012, 2012:48 doi:10.1186/1687-6180-2012-48
[12] Bayan Omar Mohammed and Siti Mariyam Shamsuddin,(2011). Feature Discretization for Individuality Representation in Twins Handwritten Identification, Journal of Computer Science 7 (7): 1080-1087, 2011, ISSN 1549-3636, Science Publications.
[13] ID. Trier and AK. Jain. (1996). Feature Extraction Methods for Character Recognition- A Survey,” Pattern Recognition, vol. 29, no. 4, 641- 662.
[14] H. Takahashi (1991). A Neural Net OCR using geometrical and zonal pattern features. In Proc. 1th. Conf. Document Analysis and Recognition, 821-828.
[15] K. Azmi, R. Kabir and E. Badi ,(2003). Recognition Printed Letters wit Zonong Features. Iran Computer Group, vol. 1, 29-37.
[16] R.Muralidharan,C.Chandrasekar,(2011) . Object Recognition using SVM-KNN based on Geometric Moment Invariant , International Journal of Computer Trends and Technology , ISSN: 2231-2803 , pp. 215-219.
