Latent Fingerprint Segmentation with Adaptive ... - Semantic Scholar

Latent Fingerprint Segmentation with Adaptive Total Variation Model Jiangyang Zhang1 1Ming

Rongjie Lai2

C.-C. Jay Kuo1

2Department of Mathematics Hsieh Department of Electrical Engineering University of Southern California

INTRODUCTION Fingerprint Segmentation Foreground (Fingerprint) Background (Noise)

Three types of Fingerprint Images 1 Rolled

2 Plain

3 Latent 1

LATENT FINGERPRINT SEGMENTATION THE CHALLENGE

Six types of Structured Noise Line

Arch

Text

Stain

Speckle

Other

Signal Fingerprint (Weak)