Irrelevant Variability Normalization via Hierarchical Deep Neural Networks for Online Handwritten Chinese Character Recognition

This paper presents a novel irrelevant variability normalization (IVN) approach via hierarchical deep neural networks (HDNNs) and prototype-based classifier for online handwritten Chinese character recognition. The recent insight of deep neural network (DNN) is the deep architecture with large training data can bring the best performance in many research areas. The architecture design of our proposed hierarchical deep neural networks focuses on both "depth" and "width" of artificial neural network. Specifically for the multivariate regression, HDNN consists of multiple subnets, which is empirically more powerful than DNN. In this work, HDNN is adopted as a nonlinear feature transform to normalize the feature vector of handwritten samples with irrelevant variabilities to a target prototype. The effectiveness of proposed method is verified on a Chinese handwriting recognition task. Furthermore, we have an very interesting observation that DNN-based IVN can not even bring performance gain over the prototype-based classifier while HDNN-based IVN yields significant improvements of recognition accuracy.

[1]  Geoffrey E. Hinton.  A Practical Guide to Training Restricted Boltzmann Machines , 2012, Neural Networks: Tricks of the Trade.

[2]  Richard M. Schwartz,et al.  A compact model for speaker-adaptive training , 1996, Proceeding of Fourth International Conference on Spoken Language Processing. ICSLP '96.

[3]  Tara N. Sainath,et al.  Deep Neural Networks for Acoustic Modeling in Speech Recognition , 2012 .

[4]  Jun Du,et al.  An Experimental Study on Speech Enhancement Based on Deep Neural Networks , 2014, IEEE Signal Processing Letters.

[5]  Jun Du,et al.  An Irrelevant Variability Normalization Based Discriminative Training Approach for Online Handwritten Chinese Character Recognition , 2013, 2013 12th International Conference on Document Analysis and Recognition.

[6]  Fei Yin,et al.  Online and offline handwritten Chinese character recognition: Benchmarking on new databases , 2013, Pattern Recognit..

[7]  Richard M. Schwartz,et al.  Discriminatively Trained Region Dependent Feature Transforms for Speech Recognition , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.

[8]  Robert M. Gray,et al.  An Algorithm for Vector Quantizer Design , 1980, IEEE Trans. Commun..

[9]  Yongqiang Wang,et al.  A Study of Designing Compact Recognizers of Handwritten Chinese Characters Using Multiple-Prototype Based Classifiers , 2010, 2010 20th International Conference on Pattern Recognition.

[10]  Masaki Nakagawa,et al.  'Online recognition of Chinese characters: the state-of-the-art , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Jin Chen,et al.  Applying Discriminatively Optimized Feature Transform for HMM-based Off-Line Handwriting Recognition , 2012, 2012 International Conference on Frontiers in Handwriting Recognition.

[12]  Qiang Huo,et al.  A study of a new misclassification measure for minimum classification error training of prototype-based pattern classifiers , 2008, 2008 19th International Conference on Pattern Recognition.

[13]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

[14]  Zhi-Jie Yan,et al.  A study of an irrelevant variability normalization based discriminative training approach for LVCSR , 2011, 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[15]  Zhen-Long Bai,et al.  A study on the use of 8-directional features for online handwritten Chinese character recognition , 2005, Eighth International Conference on Document Analysis and Recognition (ICDAR'05).

[16]  Hermann Ney,et al.  Writer Adaptive Training and Writing Variant Model Refinement for Offline Arabic Handwriting Recognition , 2009, 2009 10th International Conference on Document Analysis and Recognition.

[17]  Jun Du,et al.  A discriminative linear regression approach to OCR adaptation , 2012, Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012).

[18]  Mark J. F. Gales,et al.  Maximum likelihood linear transformations for HMM-based speech recognition , 1998, Comput. Speech Lang..

[19]  Jun Du,et al.  Designing compact classifiers for rotation-free recognition of large vocabulary online handwritten Chinese characters , 2012, 2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[20]  Yongqiang Wang,et al.  Building compact recognizers of handwritten Chinese characters using precision constrained Gaussian model, minimum classification error training and parameter compression , 2011, International Journal on Document Analysis and Recognition (IJDAR).

[21]  Kaizhu Huang,et al.  Pattern Field Classification with Style Normalized Transformation , 2011, IJCAI.

[22]  Cheng-Lin Liu,et al.  Style transfer matrix learning for writer adaptation , 2011, CVPR 2011.

[23]  Jun Du,et al.  A discriminative linear regression approach to adaptation of multi-prototype based classifiers and its applications for Chinese OCR , 2013, Pattern Recognit..

[24]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[25]  Bin Ma,et al.  Irrelevant variability normalization in learning HMM state tying from data based on phonetic decision-tree , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[26]  Tara N. Sainath,et al.  Deep Neural Networks for Acoustic Modeling in Speech Recognition: The Shared Views of Four Research Groups , 2012, IEEE Signal Processing Magazine.

[27]  Qiang Huo,et al.  Confidence guided progressive search and fast match techniques for high performance Chinese/English OCR , 2002, Object recognition supported by user interaction for service robots.