Signature recognition

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

On-line pen input signature verifier PPI (pen-position/pen-pressure/pen-inclination)

Abstract: An algorithm is proposed for pen-input online signature verification, incorporating pen-position, pen-pressure and pen-inclination trajectories. The algorithm considers the writer's signature as a trajectory of pen-position, pen-pressure and pen-inclination which evolves over time, so that it is dynamic and biometric. Since the algorithm uses pen-trajectory information, it naturally needs to incorporate stroke number (number of pen-ups/pen-downs) variations as well as shape variations. The proposed scheme first generates templates from several authentic signatures of individuals. In the verification phase, the scheme computes a distance between the template and an input trajectory. Care needs to be taken in computing the distance function because: (i) the length of a pen input trajectory may be different from that of a template even if the signature is genuine; (ii) the number of strokes of a pen input trajectory may be different from that of the template, i.e., the number of pen-ups/pen-downs obtained may differ from that of the template even for an authentic signature. If the computed distance does not exceed a threshold value, the input signature is predicted to be genuine, otherwise it is predicted to be forgery. Preliminary experimental results look encouraging.

Fusion of fingerprint, palmprint and hand geometry for an efficient multimodal person authentication system

Abstract: Biometric framework is effectively emerging in different commercial ventures for the past couple of years, and it is continuing to offer higher security features for access control system. Numerous sorts of unimodal biometric frameworks have been produced. But, these frameworks are only able to provide low to medium range of security features. As a result, for higher security features, the composite of two or more unimodal biometrics (different modalities) is required. In this paper, we propose a multimodal biometric framework for person authentication by incorporating three distinct modalities, such as, Fingerprint, palmprint and hand geometry. Unique fingerprint recognition is carried out utilizing Histogram of oriented gradients (HOG), palmprint recognition is performed utilizing combination of principal component analysis (PCA) and linear discriminant analysis (LDA) where as hand geometry recognition is accomplished utilizing Harris corner detection algorithm. The three modalities are consolidated and the fusion is applied at the feature level for both verification and identification. Integrating several biometric traits increases recognition performance and decreases fraudulent access. The identity established by this system is more reliable and robust than the identity established by individual biometric systems. The classification is performed using a matching technique such as support vector machine (SVM). The system was tested on the database of 50 persons. The experimental results revealed that the designed system attains an excellent recognition rate and offer more security than unimodal biometric-based system. The recognition performance of this recognition system is computed by means of False Acceptance Rate (FAR), False Rejection Rate (FRR) and Recognition Rate (RR).

Simple and Effective Techniques for Skew Correction, Slant Correction and Core-Region Detection for Cursive Word Recognition

Abstract: For the past decades, the advancement in the field of Image Processing has been paving a profound way in digital treatment of Human written data. Handwriting Recognition, a subset, is now a major research area to study as it is providing a mean for automatic processing of large volumes of data in reading and office automation. Intelligent word recognition systems which are used in processing important documents like bank cheques, old scripts are the need of the hour. Through this paper we present a new approach for Cursive word and Signature recognition. We propose Core-region detection technique which enables us to identify the crucial features of the hand written signatures by the extracting ’Ascenders and Descenders’. Skew and Slant corrections, if needed, are performed as preprocessing steps. A significant reduction in computation complexity has been observed than the previous attempts of researchers in detection of core-region.

Human Face Recognition

Abstract: scale invariant feature transform) being a feature extraction algorithm was initially used for object recognition, pattern recognition. But from past few years this algorithm is frequently being used for human face recognition based on feature extraction and matching. The twist here is that between feature extraction and its recognition I have used fuzzy clustering to create clusters so that we can localize the areas for feature point detection. As this recognition is based on a GUI system whose recognition rate is almost 99% - 100% for face recognition and feature detection.

A Biometric Identity Based Signature Scheme with Convenient Verification

Abstract: We describe an identity based signature scheme that uses biometric data to construct the public key. A biometric reading provided by the alleged signer would be enough to verify the signature. We make use of fractal transform and entropy arrangement algorithm to generate the public key string from a biometric measurement. We then make use of Shamir scheme [1] to perform signing and verification. Finally, we describe two possible attacks on this system and suggest ways to combat it. The result of experiments have shown that the scheme could satisfy the practical applies.