A Multichannel Approach to Fingerprint Classification
IEEE Transactions on Pattern Analysis and Machine Intelligence
IEEE Transactions on Neural Networks
Review article: Review of pulse-coupled neural networks
Image and Vision Computing
Hi-index | 0.00 |
This paper proposes a novel fingerprint classification method. It uses an SPCNN (Simplified Pulse Coupled Neural Network) to estimate directional image of fingerprint, and quantizes them to obtain fingerprint vector. Then, a fully trained LVQ (Learning Vector Quantization) neural network is used as classifier for the fingerprint vector to determine the corresponding fingerprint classification. Experiments show this proposed method is robust and has high classification accuracy.