PERBANDINGAN ALGORITMA SUM OF SQUARED DIFFERENCE (SSD) DAN OPTIMISED SUM OF ABSOLUTE DIFFERENCE (OSAD) UNTUK PENGENALAN SIMBOL PADA CITRA EKSPRESI MATEMATIKA TERCETAK

Aditya Wikan Mahastama

Abstract


The process of recognising printed mathematical expression consists of two general processes: symbol recognition  and structural analysis which reads the expression structure and the relation between symbols within the structure, provide adequate labeling to preserve it, and finally be useful to reconstruct the whole expression.

This research focused on comparing the effectivity of two correlation-based pattern matching algorithms: the well-established Sum of Squared Difference (SSD) and the relatively new Optimized Sum of Absolute Difference (OSAD) which is based on average face rather than mathing all trained symbols. Which one was better will be used on the next stage of a research on converting printed mathematical expression into LaTeX notation.. 

Results obtained from tests conducted shown that (1) the accuracy of OSAD is 85.6% and SSD is 70,6% on various test symbols, and (2) the usage of average face in OSAD which was proposed to reduce time needed for matching, proven to have no negative effects on its matching performance. Therefore OSAD is well ahead in processing time and accuracy than SSD.


Keywords


Pattern recognition, symbol recognition, OSAD, SSD

Full Text:

PDF

References


Alvaro, F., Sanchez, J-A., & Benedi, J-M. (2011). Recognition of Printed Mathematical Expressions Using Two-dimensional Stochastic Context-Free Grammars. 2011 International Conference on Document Analysis and Recognition.

Bluche, T. (2010). Mathematical Formula Recognition using Machine Learning Techniques. Tesis, University of Oxford.

Chan, K., & Yeung, D. (2000). Mathematical expression recognition: a survey. International Journal on Document Analysis and Recognition, 3, 3-15.

Dawoud, N. N., Samir, B. B., & Janier, J. (2011). Fast Template Matching Method Based Optimized Sum of Absolute Difference Algorithm for Face Localization, International Journal of Computer Applications (0975–8887), Vol. 18–No.8.

Eto, Y., & Suzuki, M. (2001). Mathematical formula recognition using virtual link network. Washington, D.C.: International Conference on Document Analysis and Recognition, 762–767.

Finlayson, G. D., Qiu, G., & Qiu, M. (1999). Contrast Maximizing and Brightness Preserving Color to Grayscale Image Conversion.

Mahastama, A.W., & Tamatjita, E. N. (2008). Optical Character Recognition dengan Algoritma OCRCHIE. Konferensi Nasional Sistem Informasi (KNSI), Yogyakarta.

Martin, J., & Crowley, J.L. (1995). Comparison of Correlation Techniques, Conference on Intelligent Autonomous Systems (IAS) 1995, Karlsruhe.

Prusa, D., & Hlavac, V. (2007). Mathematical formulae recognition using 2d grammars, International Conference on Document Analysis and Recognition, 2, 849–853.

Ranger, J., Wang, F., & Cheng, H. (2012). Optical Character Recognition of Printed Mathematical Symbols using A Hierarchical Classifier, The 2012 International Conference on Image Processing, Computer Vision, and Pattern Recognition.




DOI: http://dx.doi.org/10.21460/inf.2016.121.457

Refbacks

  • There are currently no refbacks.


Bookmark and Share