ER2: an intuitive similarity measure for on-line signature verification

ER² (extended r-squared) is proposed as a similarity measure for on-line signature verification. SLR (simple linear regression) defines R² as a measure of goodness-of-fit. We observed that R²is a good similarity measure for 1-dimensional sequences. However, many kinds of sequences are multidimensional, such as on-line signature sequences, 2D curves, etc. Therefore, we extend R² to ER² for multidimensional sequence matching. Coupled with optimal alignment, ER² outperforms DTW-based curve matching on on-line signature verification.