Off-line writer identification using Gaussian mixture models

Writer identification is the task of determining the author of a sample handwriting from a set of writers. In this paper, we propose Gaussian Mixture Models (GMMs) to address the task of off-line, text independent writer identification of text lines. The resulting system is compared to a system that uses a Hidden Markov Model (HMM) based approach. While the GMM based system is conceptually much simpler and faster to train than the HMM based system, it achieves a significantly higher writer identification rate of 98.46% on a data set of 4,103 text lines coming from 100 writers.

[1]  Horst Bunke,et al.  The IAM-database: an English sentence database for offline handwriting recognition , 2002, International Journal on Document Analysis and Recognition.

[2]  Samy Bengio,et al.  Torch: a modular machine learning software library , 2002 .

[3]  Horst Bunke,et al.  A Set of Novel Features for Writer Identification , 2003, AVBPA.

[4]  Lambert Schomaker,et al.  Automatic writer identification using connected-component contours and edge-based features of uppercase Western script , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  Sung-Hyuk Cha,et al.  MULTIPLE FEATURE INTEGRATION FOR WRITER VERIFICATION , 2004 .

[6]  Thierry Paquet,et al.  A writer identification and verification system , 2005, Pattern Recognit. Lett..

[7]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[8]  Tieniu Tan,et al.  Personal identification based on handwriting , 2000, Pattern Recognit..

[9]  Subhash C. Bagui,et al.  Combining Pattern Classifiers: Methods and Algorithms , 2005, Technometrics.

[10]  Lambert Schomaker,et al.  Automatic writer identification using fragmented connected-component contours , 2004, Ninth International Workshop on Frontiers in Handwriting Recognition.

[11]  Réjean Plamondon,et al.  Automatic Signature Verification: The State of the Art - 1989-1993 , 1994, Int. J. Pattern Recognit. Artif. Intell..

[12]  Vassilis Anastassopoulos,et al.  Morphological waveform coding for writer identification , 2000, Pattern Recognit..

[13]  Réjean Plamondon,et al.  Automatic signature verification and writer identification - the state of the art , 1989, Pattern Recognit..

[14]  Horst Bunke,et al.  Using a Statistical Language Model to Improve the Performance of an HMM-Based Cursive Handwriting Recognition System , 2001, Int. J. Pattern Recognit. Artif. Intell..

[15]  Antoine Manzanera,et al.  Improved low complexity fully parallel thinning algorithm , 1999, Proceedings 10th International Conference on Image Analysis and Processing.

[16]  Thierry Paquet,et al.  Handwriting analysis for writer verification , 2004, Ninth International Workshop on Frontiers in Handwriting Recognition.

[17]  Douglas A. Reynolds,et al.  Speaker identification and verification using Gaussian mixture speaker models , 1995, Speech Commun..

[18]  Graham Leedham,et al.  Writer identification using innovative binarised features of handwritten numerals , 2003, Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings..

[19]  Bernadette Dorizzi,et al.  Rejection measures for handwriting sentence recognition , 2002, Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition.

[20]  Frédéric Bimbot,et al.  A comparative evaluation of variance flooring techniques in HMM-based speaker verification , 1998, ICSLP.

[21]  Sargur N. Srihari,et al.  Individuality of handwritten characters , 2003, Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings..

[22]  Horst Bunke,et al.  Off-line handwriting identification using HMM based recognizers , 2004, ICPR 2004.

[23]  Douglas A. Reynolds,et al.  Speaker Verification Using Adapted Gaussian Mixture Models , 2000, Digit. Signal Process..