Improved Fuzzy Entropy for Signature Recognition

Abstract

Distinguishing different time series, which are determinant or stochastic, is an important task in signal processing. In this work, a correlation measure constructs Correlation Fuzzy Entropy (CFE) to discriminate Chaos and stochastic series. CFE based on the modulus of polar coordinates eliminates the effects of argument with a projection inequality. It can be employed to distinguish chaotic signals from ARIMA series with different noises. The CFE was verified by analyzing two available online signature databases MCYT-100 and SVC2004.