Wavelet Transform and Invariant Moments Based Off-line Handwritten Tibetan Character Recognition*

摘要

The approximation coefficient matrix of the 1-level Daubechies wavelet transform, combined with the Hu set of invariant moments, is proposed as the features of the off-line handwritten Tibetan consonant letters recognition. And an optimal coefficient, is multiplied to the Hu set, of invariant moments before concatenating it to the vector transformed from the approximation coefficient matrix. The experiments show that the proposed features improve the recognition rate of off-line handwritten Tibetan consonants evidently.