Fingerprint ROI segmentation based on deep learning

This paper presents a novel method for fingerprint ROI (region of interest) segmentation using Deep learning technique — Convolutional Neural Networks. Experimental results, obtained using a publicly available test database of 200 fingerprint images in two variations — with and without Gaussian noise, demonstrate that this method is competitive with Fourier coefficients and NN based method for fingerprint images without noise, while it significantly outperforms it, in all three figures of merit, for fingerprint images with noise.