论文信息 - Reconocimiento de dígitos manuscritos usando la transformada wavelet continua en 2 dimensiones y redes neuronales

Reconocimiento de dígitos manuscritos usando la transformada wavelet continua en 2 dimensiones y redes neuronales

In this paper we present a preprocessing method for handwritten numerals recognition, based on a two dimensional continuous wavelet transform. We use the wavelet transformed digits to input into a multilayer feed forward neural network, trained with backpropagation algorithm. Our preliminary results are promising.

Ana M. C. Ruedin | Leticia Seijas | Diego J. Romero

[1] Simon Haykin,et al. Neural Networks: A Comprehensive Foundation , 1998 .

[2] Ezequiel López-Rubio,et al. A principal components analysis self-organizing map , 2004, Neural Networks.

[3] Hong Yan,et al. A nonlinear neural network model of mixture of local principal component analysis: application to handwritten digits recognition , 2001, Pattern Recognit..

[4] Anders Krogh,et al. Introduction to the theory of neural computation , 1994, The advanced book program.

[5] Sung-Bae Cho,et al. Self-Organizing Map with Dynamical Node Splitting: Application to Handwritten Digit Recognition , 1997, Neural Computation.

[6] Yuan Yan Tang,et al. Multiresolution recognition of unconstrained handwritten numerals with wavelet transform and multilayer cluster neural network , 1996, Pattern Recognit..

[7] Jean-Pierre Antoine,et al. Two-dimensional directional wavelets and the scale-angle representation , 1996, Signal Process..

[8] Andrew F. Laine,et al. Wavelet descriptors for multiresolution recognition of handprinted characters , 1995, Pattern Recognit..

[9] Enrique Carlos Segura,et al. Un clasificador neuronal que explica sus respuestas: aplicación al reconocimiento de dígitos manuscritos , 2003 .