Numeral Recognition Based on Hierarchical Overlapped Networks

This paper describes our investigation into the neural gas (NG) algorith m and the hierarchical overlapped network architecture (HONG), which has been built by retaining the essence of t h riginal NG algorithm. By defining an implicit ranking scheme, the NG algorithm is made to run faster in its s equential implementation. The HONG network generates multiple classifications for every sample data presented as confidence values. These confidence values are combined to obtain the final classification of the HONG architecture. The proposed architecture is tested on the NIST SD3 database, which contains real world handwritten numerals wi th high variations, and an excellent recognition rate is consequently obtained.

[1]  Teuvo Kohonen,et al.  The self-organizing map , 1990, Neurocomputing.

[2]  Hong Yan,et al.  Handwritten digit recognition by neural 'gas' model and population decoding , 1998, 1998 IEEE International Joint Conference on Neural Networks Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98CH36227).

[3]  Thomas Martinetz,et al.  'Neural-gas' network for vector quantization and its application to time-series prediction , 1993, IEEE Trans. Neural Networks.

[4]  Michael D. Garris,et al.  Design, Collection, and Analysis of Handwriting Sample Image Databases , 1994 .

[5]  Ponnuthurai Nagaratnam Suganthan,et al.  Multiple HONG network fusion by fuzzy integral , 1999, ICONIP'99. ANZIIS'99 & ANNES'99 & ACNN'99. 6th International Conference on Neural Information Processing. Proceedings (Cat. No.99EX378).

[6]  Sandro Ridella,et al.  On the importance of sorting in "neural gas" training of vector quantizers , 1997, Proceedings of International Conference on Neural Networks (ICNN'97).

[7]  Wan-Chi Siu,et al.  Fast sequential implementation of "neural-gas" network for vector quantization , 1998, IEEE Trans. Commun..

[8]  Stefano Ferrari,et al.  Image reconstruction using improved "neural-gas" , 1996 .

[9]  T. M. Ha,et al.  DESIGN , IMPLEMENTATION , AND TESTING OF PERTURBATION METHOD FOR HANDWRITTEN NUMERAL RECOGNITION , 1996 .

[10]  Minyue Fu,et al.  Application of neural "gas" model in image compression , 1998, 1998 IEEE International Joint Conference on Neural Networks Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98CH36227).

[11]  Joachim M. Buhmann,et al.  An Annealed "Neural Gas" Network for Robust Vector Quantization , 1996, ICANN.

[12]  Michiharu Maeda,et al.  Destructive Fuzzy Modeling Using Neural Gas Network , 1997 .