A Study of Moment Based Features on Handwritten Digit Recognition
Open Access
- 1 January 2016
- journal article
- research article
- Published by Hindawi Limited in Applied Computational Intelligence and Soft Computing
- Vol. 2016, 1-17
- https://doi.org/10.1155/2016/2796863
Abstract
Handwritten digit recognition plays a significant role in many user authentication applications in the modern world. As the handwritten digits are not of the same size, thickness, style, and orientation, therefore, these challenges are to be faced to resolve this problem. A lot of work has been done for various non-Indicscripts particularly, in case ofRoman, but, in case ofIndicscripts, the research is limited. This paper presents a script invariant handwritten digit recognition system for identifying digits written in five popular scripts of Indian subcontinent, namely,Indo-Arabic,Bangla,Devanagari,Roman, andTelugu. A 130-element feature set which is basically a combination of six different types of moments, namely, geometric moment, moment invariant, affine moment invariant, Legendre moment, Zernike moment, and complex moment, has been estimated for each digit sample. Finally, the technique is evaluated onCMATERand MNIST databases using multiple classifiers and, after performing statistical significance tests, it is observed that Multilayer Perceptron (MLP) classifier outperforms the others. Satisfactory recognition accuracies are attained for all the five mentioned scripts.Keywords
Funding Information
- University with potential for excellence
This publication has 23 references indexed in Scilit:
- Offline Script Identification from multilingual Indic-script documents: A state-of-the-artComputer Science Review, 2015
- Efficient Handwritten Digit Recognition based on Histogram of Oriented Gradients and SVMInternational Journal of Computer Applications, 2014
- A genetic algorithm based region sampling for selection of local features in handwritten digit recognition applicationApplied Soft Computing, 2012
- "Kannada, Telugu and Devanagari Handwritten Numeral Recognition with Probabilistic Neural Network: A Script Independent Approach"International Journal of Computer Applications, 2011
- Arabic handwritten digit recognitionInternational Journal on Document Analysis and Recognition (IJDAR), 2008
- Simple Method for High-Performance Digit Recognition Based on Sparse CodingIEEE Transactions on Neural Networks, 2008
- Handwritten Bangla numeral recognition system and its application to postal automationPattern Recognition, 2007
- Handwritten digit recognition: benchmarking of state-of-the-art techniquesPattern Recognition, 2003
- Databases for recognition of handwritten Arabic chequesPattern Recognition, 2003
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998