Signature recognition

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

Towards an information theoretic framework for object recognition

Abstract: We propose a model for rate-constrained pattern recognition problems, and present single-letter information bounds governing the conditions under which asymptotically error-free recognition is possible. The bounds depend on the statistics of the training and testing data, the number of pattern classes, and the rates of the codes used by the recognition system to internalize the data.

An effective stroke feature selection method for online signature verification

Abstract: Signature is a behavioral trait of an individual and forms a special class of handwriting in which legible letters or words may not be exhibited. Dynamic signature verification (DSV) uses the behavioral biometrics of a hand-written signature to confirm the identity of a computer user. Signature verification technology requires primarily a digitizing tablet and a special pen connected to the Universal Serial Bus Port (USB port) of a computer. An individual can sign on the digitizing tablet using the special pen regardless of his signature size and position. The signature is characterized as pen-strokes consisting x, y coordinates and the data will be stored in the signature database. In this paper, I am describing the various feature of dynamic signature by considering their spatial and time domain characteristics. Individual strokes are identified by finding the points where there is a decrease in pen tip pressure, decrease in pen velocity, and rapid change in pen angle.

Visual keyword recognition using hidden Markov models

Abstract: An algorithm for robust machine recognition of keywords embedded in a poorly printed document is presented. For each keyword, two statistical models, named hidden Markov models (HMMs), are created for representing the actual keyword and all the other extraneous words, respectively. Dynamic programming is then used for matching an unknown input word with the two models and making a maximum likelihood decision. Both the 1D and pseudo-2D HMM approaches are proposed and tested. The 2D models are shown to be general enough in characterizing printed words efficiently. These pseudo-2D HMMs facilitate an elastic matching property in both the horizontal and vertical directions, which makes the recognizer not only independent of size and slant but also tolerant of highly deformed and noisy words. The system is evaluated on a synthetically created database. Recognition accuracy of 99% is achieved when words in testing and training sets are in the same font size, and 96% is achieved when they are in different sizes. In the latter case, the 1D HMM achieves only a 70% accuracy rate. >

A cepstral based speaker recognition system

Abstract: Development of a fixed limited vocabulary automatic speaker recognition system is described. The operation of the system is based on extraction of ceptstral features from single isolated word utterances by various speakers. A dynamic time warping algorithm is used in the comparison stage to bring the feature vectors being compared into time alignment. A nearest neighbor rule is used to determine the identity of the speaker. >

A Framework for Integrating Multimodal Biometrics with Digital Forensics

Abstract: Multimodal biometrics represents various categories of morphological and intrinsic aspects with two or more computerized biological characteristics such as facial structure, retina, keystrokes dynamics, voice print, retinal scans, and patterns for iris, facial recognition, vein structure, scent, hand geometry, and signature recognition. The objectives of Digital Forensics (DF), on the other hand, is to inspect digital media in a forensically sound manner with the essence of identifying, discovering, recovering, analysing the artifacts and presenting facts and suggestions about the discovered information to any court of law or civil proceedings. Because the accuracy of biometric indicators may rarely be investigated during a digital forensic investigation processes, integrating digital forensics with multimodal biometrics can enable effective digital forensic investigations on multiple captured physiological and behavioural characteristics. This paper, therefore, presents a self-adaptive approach for integrating digital forensics with multimodal biometrics. This is motivated by the fact that, as of the time of writing this paper, there is lack of effective and standardised methods for performing digital investigation across multimodal biometric indicators. In addition, there are also no proper digital forensic biometric management strategies in place. For this reason, to enable effective digital investigations on multiple captured physiological and behavioural characteristics, this paper aims at proposing a framework that is meant to facilitate the integration of DF and multimodal biometrics. The framework is also meant to enhance the analysis of potential digital evidence during investigations. Integrating multimodal biometrics and digital forensics using the proposed framework gives a promising approach to add value especially in enforcing security measures in different systems as well as a restricting factor to unauthorized access key discoveries. The integration of digital forensics with multimodal biometrics is the main focus of this paper.