In this paper, the different aspects and problems of signature recognition and verification
are discussed based on off-line analysis methods. It is shown, that on-line analysis
methods perform usually better than off-line methods. After discussing
some of the main issues of off-line analysis, methods for extracting dynamic information
from static images are introduced. Afterwards, a solution is presented for recovering
a narrow set of dynamic features, making it possible, to use them as an input for
dynamic signature recognition and verification methods.