Fingerprint classification using convolutional neural networks and ridge orientation images
John M. Shrein · 2017
Deep learning is currently popular in the field of computer vision and pattern recognition. Deep learning models such as Convolutional Neural Networks (CNNs) have been shown to be extremely effective at solving computer vision problems with state-of-the-art results. This work uses the CNN model to achieve a high degree of accuracy on the problem of fingerprint image classification. It will also be shown that effective image preprocessing can greatly reduce the dimensionality of the problem, allowing fast training times without compromising classification accuracy, even in networks of moderate depth. The proposed approach has achieved 95.9% classification accuracy on the NIST-DB4 dataset with zero sample rejection.