Scattering representation in Malayalam character recognition

Feature extraction is the process of mapping input signal to informative representation that can easily be handled by the classifier systems to build decision boundary in between the participating pattern classes. Scattering representation build invariant signal representation by applying a cascade of wavelet decompositions and complex modulus, followed by low-pass filtering. The objective of this paper is to analyze the performance of scattering representation over Malayalam character recognition process. Malayalam character recognizers built from image pixel features and the features extracted from scattering network are tested over real world document images. Soft-max Regression classifier is utilized for building the classification models. Scattering representation based recognition system could achieve a 2% increase in recognition accuracy compared to image pixel value based features.

[1]  C. V. Jawahar,et al.  Towards a Robust OCR System for Indic Scripts , 2014, 2014 11th IAPR International Workshop on Document Analysis Systems.

[2]  R. Ramanathan,et al.  Robust Feature Extraction Technique for Optical Character Recognition , 2009, 2009 International Conference on Advances in Computing, Control, and Telecommunication Technologies.

[3]  C. V. Jawahar,et al.  Empirical Evaluation of Character Classification Schemes , 2009, 2009 Seventh International Conference on Advances in Pattern Recognition.

[4]  Stéphane Mallat,et al.  Group Invariant Scattering , 2011, ArXiv.

[5]  K. P. Soman,et al.  Novel SVD Based Character Recognition Approach for Malayalam Language Script , 2013, ISI.

[6]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[7]  V. K. Govindan,et al.  Character recognition - A review , 1990, Pattern Recognit..

[8]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[9]  Bidyut Baran Chaudhuri,et al.  Indian script character recognition: a survey , 2004, Pattern Recognit..

[10]  Hao Shen,et al.  Texture Retrieval via the Scattering Transform , 2015, ArXiv.

[11]  Bill Triggs,et al.  Histograms of oriented gradients for human detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[12]  S. Mallat,et al.  Invariant Scattering Convolution Networks , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  P. M. Dhanya,et al.  Multiple Classifier System for Offline Malayalam Character Recognition , 2015 .

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

[15]  M AbdulRahiman.  Printed Malayalam Character Recognition Using Back-propagation Neural Networks , 2009 .

[16]  Mariusz Kleć,et al.  Unsupervised Feature Pre-training of the Scattering Wavelet Transform for Musical Genre Recognition , 2014 .