Signature recognition

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

On the use of edge features and exponential decaying number of nodes in the hidden layers for handwritten signature recognition

Abstract: Handwritten signatures are playing an important role in finance, banking and education and more because it is considered the “seal of approval” and remains the most preferred means of authentication. In this paper, an offline handwritten signature authentication algorithm is proposed using the edge features and deep feedforward neural network (DFNN). The number of hidden layers in DFNN is configured to be at least one layer and more. In this paper, an exponential decaying number of nodes in the hidden layers was proposed to achieve better recognition rate with reasonable training time. Of the six edge algorithms evaluated, Roberts operator and Canny edge detectors were found to produce better recognition rate. Results showed that the proposed exponential decaying number of nodes in the hidden layers outperform other structure. However, more training data was required so that the proposed DFNN structure could have more efficient learning.

Off-line signature verification using curve fitting algorithm with neural networks

Abstract: As signature is widely used as a means of personal verification, it is necessary for an automatic verification system. Offline and Online are two methods of verification based on the application. Online systems use dynamic information of a signature captured at the time the signature is made. Offline systems work on the scanned image of a signature. Processing Off-line is complex due to the absence of stable dynamic characteristics and also due to highly stylish and unconventional writing styles. A simple and a reliable system has to be designed which should detect various types of forgeries. Hence this paper proposes architecture for off-line signature verification. Our approach makes use of runtime signature instead of scanned images for recognition. This Offline verification of signatures uses a set of shape based geometric features and more importantly focuses on the distance based parameters such as the continuity of the signature and matching of the curves of the signatures generated by the critical points of the respective signature by analyzing the polynomial equation. Curve fitting and the analyzing of polynomial equations is one of the least explored topics till date but yet very efficient and hence we have implement this novel technique.

A Comparative Study of User Dependent and Independent Accelerometer-Based Gesture Recognition Algorithms

Abstract: In this paper, we introduce an evaluation of accelerometer-based gesture recognition algorithms in user dependent and independent cases. Gesture recognition has many algorithms and this evaluation includes Hidden Markov Models, Support Vector Machine, K-nearest neighbor, Artificial Neural Net-work and Dynamic Time Warping. Recognition results are based on acceleration data collected from 12 users. We evaluated the algorithms based on the recognition accuracy related to different number of gestures from two datasets. Evaluation results show that the best accuracy for 8 and 18 gestures is achieved with dynamic time warping and K-nearest neighbor algorithms.

Human-Assisted Signature Recognition Based on Comparative Attributes

Abstract: This work analyzes the performance of comparative attributes labeled by humans for handwritten semi-automatic signature recognition Despite the large deployment of automatic sys-tems, humans have still an important role in many tasks relat-ed to handwritten signature recording and verification How humans can help to improve these processes is a primary aim of different research lines Comparative attributes try to ex-ploit the abilities of humans to extract discriminant infor-mation of the signatures Instead of absolute attributes (eg is this stroke vertical?), the comparative attributes offer richer responses (eg how vertical is this stroke?) In this work we present a new semi-automatic signature labeling interface in-spired by Forensic Document Examiners (FDE) Fifteen com-parative attributes have been labeled by 21 laymen, where each one carries out the labeling of 28 signatures from 130 users of the publicly available corpus BiosecurID database Through the manual labeling, a new Bio-HSL (Biometric-Handwritten Signatures Labelling) database is generated, which contains 4,968,600 signature attributes The results show that compara-tive attributes outperform absolute attributes for semi-automatic signature recognition with Equal Error Rates ranging from 55% for random comparisons to 212% for simulated forgeries