Biometric verification is becoming widely used in IT security as a trustworthy technology. Fingerprint, face and iris recognition are all based on the physical properties of an individual. It is, however, not widely known that electronic signatures also contain biometric data which can be used accurately to verify an individual’s identity. An individual’s writing habits can be obtained through a dynamic signature capture, which can then be compared with the relevant biometric data to prove whether a document is false or not. This is also widely used in financial service transactions.
Once a dynamic signature has been captured through a digital signature tablet, a number of characteristics of an individual’s handwriting are compared in order to reach a judgement, including the x-y co-ordinates of the penstrokes, length of time spent writing, speed, pressure and beginning and finishing position. These data are unique to every individual. Even if a forger can imitate the appearance of a signature, these biometric data are difficult to copy. Dynamic signature verification is more secure and reliable than verifying the appearance of a static signature.
PendoDSV SDK is a dynamic biometric verification engine. It can compare two dynamic items of data to determine their similarity, thereby being able to verify a signature.
Data sampling and verification:
When a new user is enrolled, the signer has to provide some reference signatures to the PendoDSV SDK. These signatures must be uniform and identical. Once the data has been obtained, it can be stored in a database with a unique ID in order to be used for future verification. The unique ID is used to for privacy protection purposes. The data are pre-encrypted and only decrypted before being used for verification. The entire process guarantees the safety of the signer’s data.
When the signatures are being compared, a signature from the same signer is captured on a digital pad and sent to the PendoDSV engine. The DSV SDK uses the same methods to determine the characteristics of the signature to be compared. This is then compared with previously-collected signatures from the database. Finally, the comparison results are provided using different numerical values which can be set according to the user’s requirements. The DSV SDK currently provides two levels of security, normal and high. These levels are differentiated by the strictness of the verification. Finally, the validity of the signature is decided in accordance with the results of the comparison.