Signature recognition

About: Signature recognition is a research topic. Over the lifetime, 2138 publications have been published within this topic receiving 37605 citations.

Automated verification and prevention of spoofing for biometric data

Abstract: The invention provides a method of and system for automated verification and prevention of spoofing for biometric data, such as handwritten signatures. Biometric data known to be true, such as signatures already entered by the authentic person, are recorded and compared against test biometric data, such as a test signature entered by a person seeking authentication. The test biometric data is compared against the known biometric data, and is accepted only if the test biometric data is sufficiently "close" to the known biometric data, but not so close as to indicate that known biometric data was recorded and played back for the test. The test biometric data represents a handwritten signature given contemporaneously by the person seeking access, and is verified against a set of template signatures earlier given by at least one authorized person. A set of features are extracted from both the template signatures and the test signature; comparison of these features yeilds a distance measure between the test signature and the template signature. If the distance measure is either too large or too small, the test signature is rejected. The extracted features for a set of test signatures which were accepted in the past is also recorded, and the test signature is rejected if it is identical to any of the past test signatures.

The application of signature recognition to transparent handwriting verification for mobile devices

Abstract: Purpose – The popularity of mobile devices and the evolving nature of the services and information they can delivery make them increasingly desirable targets for misuse. The ability to provide effective authentication of the user becomes imperative if protection against misuse of personally and financially sensitive information is to be provided. This paper discusses the application of biometrics to a mobile device in a transparent and continuous fashion and the subsequent advantages and disadvantages that are in contention with various biometric techniques.Design/methodology/approach – An investigation was conducted to evaluate the feasibility of utilising signature recognition, to verify users based upon written words and not signatures, thereby enabling transparent handwriting verification. Participants were required to write a number of common words, such as “hello” “sorry” and “thank you”. The ability to correctly verify against their own template and to reject impostors was then established.Findings...

Offline handwritten Devanagari word recognition: A segmentation based approach

Abstract: A novel segmentation based approach is proposed for recognition of offline handwritten Devanagari words. Stroke based features are used as feature vectors. A hidden Markov model is used for recognition at pseudocharacter level. The word level recognition is done on the basis of a string edit distance.

An efficient offline signature identification method based on Fourier Descriptor and chain codes

Abstract: This paper proposes a novel offline signature identification method based on Fourier Descriptor (FD) and Chain Codes features. Signature identification was classified into two different problems: recognition and verification. In recognition process, we used Principle Component Analysis. In verification process, we designed a multilayer feed forward artificial neural network. The main steps of constructing a signature identification system are discussed and experiments on real data sets show that the average error rate can reach 3.8%.

Biometric template protection using watermarking with hidden password encryption

Abstract: For quite a few years the biometric recognition techniques have been developed. Here, we briefly review some of the known attacks that can be encountered by a biometric system and some corresponding protection techniques. We explicitly focus on threats designed to extract information about the original biometric data of an individual from the stored data as well as the entire authentication system. In order to address security and privacy concerns, we present a biometric authentication scheme that uses two separate biometric features combined by watermark embedding with hidden password encryption to obtain a non-unique identifier of the personage. Furthermore, to present the performance of the authentication system we provide experimental results. The transformed features and templates trek through insecure communication line like the Internet or intranet in the client-server environment. Our projected technique causes security against attacks and eavesdropping because the original biometric will not be exposed anywhere in the authentication system.