Abstract
Character recognition is a branch of pattern recognition, the problem of noisy character image recognition is solved by discrete Hopfield neural network which is used as associative memory and the wavelet transform theory in the paper. The important original data is extracted from each character image, which is learning data of neural network. The character image is recognized by discrete Hopfield neural network and de-noising by the wavelet transform theory. Noisy samples of character recognition show that the method is accurate and efficient.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have