Signature Identification Menggunakan Metode Template Matching dan Fuzzy K-Nearest Neighbor

Abstract — Signature is the result of process writing a person particular nature as symbolic substance, which means symbol or mark. usually used an identifying mark person, each must have his own signature in different pattern. Because it's person's badge, Signatures now become particularly susceptible to counterfeiting and abuse that require check with pattern recognition. This research has created recognition system using methods Template Matching Fuzzy K-Nearest Neighbor help recognize The number signatures 110 two categories: original 100 data false 10 data, there were classes taken smartphone cameras. From this research, it was found best value from image size 200x200 pixels 92% class owned legible, Positive Predictive Value (PPV) 88% False Rejection Rate (FRR) 12%, k=3 on signature, 90% Negative (NPV) dan Acceptance (FAR) 10% k=9 signature. these results, could be concluded for