Security is one of the major issues in today’s world and most of us have to deal with some sort of passwords in our daily lives; but, these passwords have some problems of their own. If one picks an easy-to-remember password, then it is most likely that somebody else may guess it. On the other hand, if one chooses too difficult a password, then he or she may have to write it somewhere (to avoid inconveniences due to forgotten passwords) which may again lead to security breaches. To prevent passwords being hacked, users are usually advised to keep changing their passwords frequently and are also asked not to keep them too trivial at the same time. All these inconveniences led to the birth of the biometric field. The verification of handwritten signature, which is a behavioral biometric, can be classified into off-line and online signature verification methods. Online signature verification, in general, gives a higher verification rate than off-line verification methods, because of its use of both static and dynamic features of the problem space in contrast to off-line which uses only the static features. Despite greater accuracy, online signature recognition is not that prevalent in comparison to other biometrics. The primary reasons are: • It cannot be used everywhere, especially where signatures have to be written in ink; e.g. on cheques, only off-line methods will work. • Unlike off-line verification methods, online methods require some extra and special hardware, e.g. electronic tablets, pressure sensitive signature pads, etc. For off-line verification method, on the other hand, we can do the data acquisition with optical scanners. • The hardware for online are expensive and have a fixed and short life cycle. In spite of all these inconveniences, the use online methods is on the rise and in the near future, unless a process requires particularly an off-line method to be used, the former will tend to be more and more popular.
Kernel based K-Nearest Neighbor Method to Enhance the Performance and Accuracy of Online Signature Recognition
