ICA and BP neural network based fingerprint recognition

According to the recognition of fuzzy fingerprint and the ones with strong noise, proposed a new method which combining the ICA (Independent Component Algorithm) and BP (Back Propagation) neural network. First, using the FastICA method to extract fingerprint characteristics, then classify and recognize them by a three- layers BP neural network. This method combines the local feature extraction capability of ICA, as well as the adaptive ability and robustness of BP neural network. Experiments show that this method has a higher recognition rate of the fingerprints with strong noise. Keywords-Fingerprint recognition; strong noise; FastICA; BP neural network