Description
Signatures have been used to authenticate documents from a long time. It is a behavioral metric which is not based on the physical properties of a person such as the fingerprint or face. Authenticating an identity is very crucial today. With a lot of data used an individual's personal data can be easily compromised. With the advent of technology the number of smartphone users is increasing rapidly. The capability of these smartphones is almost limitless. A user stores a huge amount of personal data on his smartphone which includes his personal email accounts, social networking accounts, bank accounts etc. This data is vulnerable if the smartphone is not provided with application or a lock that provides access only to the user. It is very necessary to have some kind of application so that no one else except the user himself can access his personal accounts and data. This thesis involves designing a mobile application that provides security to the user's personal accounts. A user will initially train the application by putting a set of signatures. For designing this application we will be using the Dynamic Time Warping algorithm. Once the application is trained one signature will be used as a reference signature. When the user tries to log in by signing a signature, this input signature will be compared to the reference signature and if it is within a certain threshold of the reference signature the user will be able to log into the application and access his personal accounts.
