A New Offline Persian Hand Writer Recognition based on 2D-Wavelet Transforms
Автор: Keivan Borna, Vahid Hajihashemi
Журнал: International Journal of Image, Graphics and Signal Processing(IJIGSP) @ijigsp
Статья в выпуске: 9 vol.7, 2015 года.
Бесплатный доступ
All the works on writer's handwritten letters detection was based on the western languages and partly Chinese and Hindi, and there is a little study on Persian handwritten letters detection. Accordingly, in this paper a method is proposed to distinguish scanned Persian handwritten texts with image processing techniques. The proposed method assumes that the writer's handwritten are available in separate letters. The system trains with features extraction of these separate letters and then the trained system is used to detect individual handwritten among some indistinctive handwritten texts. The characteristics of our proposed method including high-speed of training in too much number of handwritten, is the content inattention and visual features considering. The results of procedures on 100 persons also admitted that the proposed method has a very good performance on Persian handwritten texts detection.
Writers' handwritten identification, handwritten, wavelet transform, nearest neighbor search
Короткий адрес: https://sciup.org/15013905
IDR: 15013905
Текст научной статьи A New Offline Persian Hand Writer Recognition based on 2D-Wavelet Transforms
Published Online August 2015 in MECS
Writers’ detection methods are divided into two general categories:
Offline Methods: In these methods, only scanned images of handwritten characters are available and the features are extracted due to the whole image or word structure. Obviously, in these methods, much dynamic information, such as pen pressure, writing speed or direction, which relate to the writing style is lost and this is more difficult than online methods. Offline or out of line methods are divided into two general groups, including text-dependent and text independent. In textdependent methods, the author should write a constant text, in order to determine his identity, but in the text independent methods, the author's identity is determined with any kind of handwritten text. Text independent methods are more complex and more useful.
Online Methods: In these methods, in addition to the visual characteristics, the dynamic information such as pen pressure, the writing priority, the writing speed, the form of the pen hits, the time when pen not lift the paper etc. is used and identification is done more accurately, due to having more information. But, these methods have limited applications, for example, these methods do not apply in identifying sign or judicial applications.
Based on the little study on Persian handwritten letters detection, in this paper a method is proposed to distinguish scanned Persian handwritten texts with image processing techniques.
The rest of this paper is organized as follows. In Section II the literature review for the problem is presented. The wavelet transform theory is described in Section III. In Section IV our proposed algorithm consisting of feature extraction and classification and the pseudo-code of it is presented. Finally, Section V is devoted to the study of the output of our algorithm and several conclusions are reported.
-
II. Literature Review
Список литературы A New Offline Persian Hand Writer Recognition based on 2D-Wavelet Transforms
- Réjean Plamondon, Christian O'Reilly, Javier Galbally, Abdullah Almaksour, éric Anquetil, Recent developments in the study of rapid human movements with the kinematic theory: Applications to handwriting and signature synthesis, Pattern Recognition Letters, Volume 35, 1 January 2014, Pages 225-235.
- M. Djioua, R. Plamondon, Studying the variability of handwriting patterns using the Kinematic Theory, Human Movement Science, Volume 28, Issue 5, October 2009, Pages 588-601.
- D. Bertolini, L.S. Oliveira, E. Justino, R. Sabourin, Texture-based descriptors for writer identification and verification, Expert Systems with Applications, Volume 40, Issue 6, May 2013, Pages 2069-2080.
- Imran Siddiqi, Nicole Vincent, Text independent writer recognition using redundant writing patterns with contour-based orientation and curvature features, Pattern Recognition, Volume 43, Issue 11, November 2010, Pages 3853-3865.
- S.M. Saad., Application of fuzzy logic and genetic algorithm in biometric text-independent writer Identification, IET Information Security.vol.5, no.1, pp.1-9, 2011.
- Golnaz Ghiasi, Reza Safabakhsh, Offline text-independent writer identification using codebook and efficient code extraction methods, Image and Vision Computing, Volume 31, Issue 5, May 2013, Pages 379-391.
- Shivram, A.; Ramaiah, C.; Porwal, U.; Govindaraju, V., "Modeling Writing Styles for Online Writer Identification: A Hierarchical Bayesian Approach," Frontiers in Handwriting Recognition (ICFHR), 2012 International Conference on , vol., no., pp.387,392, 18-20 Sept. 2012.
- Chawki Djeddi, Imran Siddiqi, Labiba Souici-Meslati, Abdellatif Ennaji, Text-independent writer recognition using multi-script handwritten texts, Pattern Recognition Letters, Volume 34, Issue 10, 15 July 2013, Pages 1196-1202.
- A.A. Brink, R.M.J. Niels, R.A. van Batenburg, C.E. van den Heuvel, L.R.B. Schomaker, Towards robust writer verification by correcting unnatural slant, Pattern Recognition Letters, Volume 32, Issue 3, 1 February 2011, Pages 449-457.
- B. Helli and M. E. Moghaddam, "A text-independent Persian Writer Identification based on Feature Relation Graph (FRG)", Pattern Recognition 43: 2199–2209 (2010).
- S. A. Medjahed, A Comparative Study of Feature Extraction Methods in Images Classification, International Journal of Image Graphics and Signal Processing (IJIGSP), 7: 16-23 (2015).
- S. Azam, F. T. Zohra, M. M. Islam, A State-of-the-art Review on Wavelet Based Image Resolution Enhancement Techniques: Performance Evaluation Criteria and Issues, International Journal of Image Graphics and Signal Processing (IJIGSP), 6: 35-46 (2014).