Abstract

Automatic Handwritten Digits Recognition (HDR) is the process of interpreting handwritten digits by machines. There are several approaches for handwritten digits recognition. In this paper we have proposed an appearance feature-based approach which process data using Histogram of Oriented Gradients (HOG). HOG is a very efficient feature descriptor for handwritten digits which is stable on illumination variation because it is a gradient-based descriptor. Moreover, linear SVM has been employed as classifier which has better responses than polynomial, RBF and sigmoid kernels. We have analyzed our model on MNIST dataset and 97. 25% accuracy rate has been achieved which is comparable with the state of the art.

References

• Hu D, Research and application of handwritten numeral recognition method, SM thesis, University of Nanchang, Nanchang, China. 2012
• Neera Saxena, Qasima Abbas Kazmi, Chandra Pal and O. P. Vyas, Employing Neocognitron Neural Network Base Ensemble Classifiers To Enhance Efficiency of Classification In Handwritten Digit Datasets. D. C. Wyld, et al. (Eds): CCSEA 2011, CS & IT 02, pp. 408–416, 2011.
• Nibaran Das, Ram Sarkar, Subhadip Basu, Mahantapas Kundu, Mita Nasipuri, Dipak Kumar Basu: A genetic algorithm based region sampling for selection of local features in handwritten digit recognition application. Appl. Soft Comput. (ASC) 12(5):1592-1606 (2012)
• Ângelo Cardoso, Andreas Wichert: Handwritten digit recognition using biologically inspired features. Neurocomputing (IJON) 99:575-580 (2013)
• N. Dalal and B. Triggs. Histograms of oriented gradients for human detection. In CVPR, 2005
• C. -C. Chang and C. -J. Lin. Libsvm: A library for support vector machines. ACM T. Intell. Syst. Technol. , 2(3):1–27, 2011
• MNIST Handwritten Digits database of New York University http://www. cs. nyu. edu/~roweis/data. html
• M. Narasimha Murty, V. Susheela Devi. An Application: Handwritten Digit Recognition, ISBN: 978-0-85729-494-4, Springer, 2011
• Michael E. Tipping, The relevance vector machine, Advancesin Neural Information Processing Systems, vol. 12, The MITPress,2000,pp. 652–658
• P. Simard, Y. LeCun, J. Denker, B. Victorri, Transformation invariance 400 in pattern recognition, tangent distance and tangent propagation, Neural Networks: Tricks of the Trade, Springer, 1998.
• Chang Liu, Tao Yan, WeiDong Zhao, et al. , Incremental Tensor Principal Component Analysis for Handwritten Digit Recognition, Mathematical Problems in Engineering, vol. 2014, Article ID 819758, 10 pages, 2014
• You Qian, Wang Xichang, Zhang Huaying, Sun Zhen, Liu Jiang, Recognition Method for Handwritten Digits Based on Improved Chain Code Histogram Feature, 3rd Int. Conf. Multimedia Technology, 2013
• B. Scholkopf, C. Burges,V. Vapnik, Extracting support data for a given task, First International Conf. Knowledge Discovery & Data Mining, AAAI Press, MenloPark, CA, 1995
• B. Scholkopf, P. Simard, A. Smola, V. Vapnik, Prior knowledge in support vector kernels, Advances in Neural Information Processing Systems, vol. 10, MITPress, 1998, pp. 640–646.
• P. Simard, Y. Le Cun,J. S. Denker, Efficient pattern recognition using a new transformation distance, in: Advances In Neural Information Processing Systems, vol. 5, Morgan Kaufmann, 1993, pp. 50–58
• Xiao-Xiao Niu, Ching Y. Suen: A novel hybrid CNN-SVM classifier for recognizing handwritten digits. Pattern Recognition (PR) 45(4):1318-1325 (2012)