On-line signature verification based on PCA (Principal Component Analysis) and MCA (Minor Component Analysis)

On-line signature verification is still an active topic in the field of biometrics. This paper proposes a novel method based on PCA (Principal Component Analysis) and MCA (Minor Component Analysis). Different from the application of PCA in other fields, both principal and minor components are used to signature verification, and MC plays a very important role. Comparing with DTW and the discriminance of Euclidean distance, the method based on PCA and MCA is better. With 1215 signatures contributed by 81 signers of which numbers of reference signatures, genuine signatures and forgeries (skilled) are 5 respectively, the EER is about 5%.