N-SVM combination and tangent vectors for handwritten alphanumeric character recognition

We propose a system for handwritten alphanumeric character recognition that is based on the learning of tangent similarities. The specific tangent vectors which constitute the a priori knowledge of each class are generated from the training data. Based on these tangent vectors, similarities with respect prototypes of classes are computed to be used as data features. In addition, we investigate the use of N-SVM combination in order to improve the error rate and reduce the runtime compared to the standard SVM. Experiments conducted on a database obtained by combining USPS and C-Cube data indicate that the proposed system gives the best performance in terms of training time and error rate.

[1]  Hassiba Nemmour,et al.  Handwritten Digit Recognition Based on A Neural – SVM Combination , 2010 .

[2]  Hang Joon Kim,et al.  Support Vector Machines for Texture Classification , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Jian-xiong Dong,et al.  Fast SVM training algorithm with decomposition on very large data sets , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Anil K. Jain,et al.  Feature extraction methods for character recognition-A survey , 1996, Pattern Recognit..

[5]  Bernhard Schölkopf,et al.  Training Invariant Support Vector Machines , 2002, Machine Learning.

[6]  Jamshid Shanbehzadeh,et al.  Persian/arabic handwritten word recognition using M-band packet wavelet transform , 2008, Image Vis. Comput..

[8]  Yann LeCun,et al.  Efficient Pattern Recognition Using a New Transformation Distance , 1992, NIPS.

[9]  Yann LeCun,et al.  Transformation Invariance in Pattern Recognition-Tangent Distance and Tangent Propagation , 1996, Neural Networks: Tricks of the Trade.

[10]  Bernard Haasdonk,et al.  Tangent distance kernels for support vector machines , 2002, Object recognition supported by user interaction for service robots.

[11]  Ching Y. Suen,et al.  Discrimination of similar handwritten numerals based on invariant curvature features , 2005, Pattern Recognit..

[12]  Sargur N. Srihari,et al.  On-Line and Off-Line Handwriting Recognition: A Comprehensive Survey , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  Adam Krzyzak,et al.  Rotation invariant feature extraction using Ridgelet and Fourier transforms , 2006, Pattern Analysis and Applications.

[14]  Hermann Ney,et al.  Adaptation in statistical pattern recognition using tangent vectors , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Chih-Jen Lin,et al.  A comparison of methods for multiclass support vector machines , 2002, IEEE Trans. Neural Networks.

[16]  Thorsten Joachims,et al.  Making large scale SVM learning practical , 1998 .