Introduction
Questions related to the identity of individuals such as “Is this the person who he or she claims to be?”, “Has this applicant been before?”, “Should this individual be given access to our system?” are asked millions of times every day by organizations in financial services, particularly banking sector.
With the increase in the number of security breaches and transaction frauds, the need for a highly secure identification and personal verification technologies is becoming apparent. In this scenario, biometric technologies stand out to be the best option that can prevent or minimize security threats effectively with a wide range of products available in the market.
Biometrics are automated methods of recognizing a person based on a physiological or behavioral characteristic: Physiological biometrics are based on measurements and data retrieved from direct measurement of a part of the human body. Fingerprint, iris-scan, retina-scan, hand geometry, and facial recognition are leading physiological biometrics. Behavioral biometrics are based on measurements and data derived from an action, and indirectly measure characteristics of the human body. Voice recognition, keystroke-scan, and signature-scan are leading behavioral biometric technologies.
Signature Recognition SYstem
Signature is a simple, concrete expression of the unique variations in human hand geometry. The way a person signs his or her name is known to be characteristic of that individual. However, no two signatures of a person are exactly identical; and are influenced by physical and emotional conditions of a subject. In addition to the general shape of the signed name, a signature recognition system can also measure pressure and velocity of the point of the stylus across the sensor pad. The variations from a typical signature also depend upon the physical and emotional state of a person. The identification accuracy of systems based on this highly behavioral biometric is reasonable but does not appear to be sufficiently high to lead to large-scale recognition. There are two approaches to identification based on signature:
· Off-line or Static - Static signature identification uses only the geometric (shape) features of a signature, where as dynamic (online) signature identification uses both the geometric (shape) features.
· Online or Dynamic – This is latest trend in SRS and , this has dynamic features such as acceleration, velocity, pressure, and trajectory profiles of the signature.
Dynamic Signature Verification system architecture
Dynamic Signature verification system consists of four subsystems:
· Data acquisition
· Signature preprocessing
· Feature extraction
· Signature verification
In the data acquisition subsystem, signatures are acquired and digitalized by a digital input pad, and the system will measure the raw data at every millisecond. A total of four channels of raw data will be measured: the sampling time t, x position, y position, and pressure p. A representative signature is shown in the below Figure. The pressure values and position are represented by the filled dot size, and the open circles indicate moments when the pen lifted up from the pad surface. Based on the four channels of raw data, the velocity, acceleration and angle signals are computed in the signature preprocessing subsystem. In addition, the dynamic signature signals are re-sampled and normalized to a standard length and missing data points interpolated before being sent to the feature extraction subsystem
Subsequently, feature information from the input dynamic signature is calculated by pre-configured feature extractors. For the training signatures, the extracted sample feature vectors are stored in the signature template database; for a test signature, the calculated feature vector is sent to the signature verification subsystem and compared against an enrolled template by a signature classifier and a match score calculated. A verification decision is made by comparing the match score with a threshold.
Security level Challenges in DSV
Improving security, or improving usability while maintaining security, is a main driver for using biometrics. Though biometrics has been successfully deployed to improve security in many applications with co-operative users. There are several security issues that have yet to be fully addressed if biometric methods are to be deployed more widely.
In generally, there are three types of forgeries in DSV system:
1) Zero-effort forgery: a random scribble or signature of another individual
2) Home-improved forgery: are made when the forger has a paper copy of the signature in possession and has ample time at home to practice and copy the signature at home. The imitation is based on just a static image of the signature.
3) Over-the-shoulder forgery: here the forger is present while the genuine signature is written. The forger does not just learn the spatial image of the signature but also the dynamic properties by observing the signing process. Combined over-the-shoulder and home improved forgeries are called skilled forgeries.
4) Professional forgery: produced by individuals those are skilled in the art of hand writing analysis.
Advantages and DisAdvantages in DSV
ADVANTAGES in DSV:
One of the advantages of this type of biometric system is the fact that signatures have been an accepted means of identity verification for centuries. This encourages a biometric technology that is easy for organizations and consumers to accept and to trust.
Another advantage of a DSV system is in the replacement of PINs or Password or keycards identifications that can be lost or stolen or forgotten is eliminated and replaced with simple signature.
· Cost of implementing DSV appears to be low end as compared with other biometric system.
· Low total error rate (about 1.5% per session).
· Identification that can be stolen, lost, or forgotten are eliminated and replaced with a simple signature.
· Forgery is detected even when the forger has managed to get a copy of the authentic signature.
· Fast and simple training only need to demonstrate the system.
· Cheap hardware.
· Little storage requirement (less than 1KB).
· Fast response (Perform 40 verification per second).
· Results do not depend on the native language of the user.
DisADVANTAGES in DSV:
DSV system will verify subjects based on the traits of their unique signature. As a result, individuals who do not sign their names in a consistent manner may have difficulty enrolling and verifying in DSV. Individuals with muscular illness and people who sign with only their initials might result in a higher false rejection rate.
Finally, signature can be affected by behavioral factors such as stress or distractions could cause vary from their normal signature sequence, so there may be some rejection happen. Another main thing is frequency-of-use-factor; a person’s signature could vary over a period of time.
Conclusion
The rule in the domain of DSV so far is the lack of normalization and, obviously, this constitutes one of their greatest disadvantages. However, compared with other identity verification systems, DSV show better reliability and performance. False Rate also less comparing to the other biometrics system, Easy to use and highly secure. Because of these reasons DSV is the perfect system to adobe in the Banking sector.
