Signature Verification Using Static and Dynamic Features
22 Nov 2004-pp 350-355
TL;DR: A signature verification algorithm based on static and dynamic features of online signature data is presented and combines the results obtained from three feature sets to attain an accuracy of 98.18%.
Abstract: A signature verification algorithm based on static and dynamic features of online signature data is presented. Texture and topological features are the static features of a signature image whereas the digital tablet captures in real-time the pressure values, breakpoints, and the time taken to create a signature. 1D – log Gabor wavelet and Euler numbers are used to analyze the textural and topological features of the signature respectively. A multi-classifier decision algorithm combines the results obtained from three feature sets to attain an accuracy of 98.18%.
Citations
More filters
TL;DR: A novel approach is explored and evaluated that takes advantage of the performance boost that can be reached through the fusion of on-line and off-line signatures and of their potential combination both in the random and skilled impostors scenarios.
104 citations
Cites background from "Signature Verification Using Static..."
...Such skilled forgeries usually lie inside the subject’s intraclass variability leading to a significant decrease of the recognition performance....
[...]
TL;DR: A novel multi-level wavelet based fusion algorithm that combines information from fingerprint, face, iris, and signature images of an individual into a single composite image that reduces the memory size, increases the recognition accuracy using multi-modal biometric features, and withstands common attacks.
34 citations
Cites methods from "Signature Verification Using Static..."
...The proposed signature verification algorithm uses 1D Gabor wavelet [5] for feature extraction [27]....
[...]
Proceedings Article•
11 Mar 2015TL;DR: An improvement of 4.7% in gender classification system is reported by the inclusion of Euler number as a feature in a feature based gender detection method.
Abstract: Recent growth of technology has also increased identification insecurity. Signature is a unique feature which is different for every other person, and each person can be identified using their own handwritten signature. Gender identification is one of key feature in case of human identification. In this paper, a feature based gender detection method has been proposed. The proposed framework takes handwritten signature as an input. Afterwards, several features are extracted from those images. The extracted features and their values are stored as data, which is further classified using Back Propagation Neural Network (BPNN). Gender classification is done using BPNN which is one of the most popular classifier. The proposed system is broken into two parts. In the first part, several features such as roundness, skewness, kurtosis, mean, standard deviation, area, Euler number, distribution density of black pixel, entropy, equi-diameter, connected component (cc) and perimeter were taken as feature. Then obtained features are divided into two categories. In the first category experimental feature set contains Euler number, whereas in the second category the obtained feature set excludes the same. BPNN is used to classify both types of feature sets to recognize the gender. Our study reports an improvement of 4.7% in gender classification system by the inclusion of Euler number as a feature.
27 citations
15 Dec 2014
TL;DR: This paper proposes several approaches to the synthesis of off-line enhanced signatures from real dynamic information, showing a performance very similar to the one offered by real signatures, even increasing their discriminative power under the skilled forgeries scenario, one of the biggest challenges of handwriting recognition.
Abstract: One of the main challenges of off-line signature verification is the absence of large databases. A possible alternative to overcome this problem is the generation of fully synthetic signature databases, not subject to legal or privacy concerns. In this paper we propose several approaches to the synthesis of off-line enhanced signatures from real dynamic information. These synthetic samples show a performance very similar to the one offered by real signatures, even increasing their discriminative power under the skilled forgeries scenario, one of the biggest challenges of handwriting recognition. Furthermore, the feasibility of synthetically increasing the enrolment sets is analysed, showing promising results.
21 citations
Cites methods from "Signature Verification Using Static..."
...Some efforts have been performed through off-line signature verification using simple methods to create static signatures [8], [9]....
[...]
TL;DR: A method to compute the Euler number of a binary digital image based on a codification of contour pixel sof the image’s shapes is described, supported through an experimental set which analyzes some digital images and their outcome to demonstrate the applicability of the procedure.
20 citations
References
More filters
TL;DR: A common theoretical framework for combining classifiers which use distinct pattern representations is developed and it is shown that many existing schemes can be considered as special cases of compound classification where all the pattern representations are used jointly to make a decision.
Abstract: We develop a common theoretical framework for combining classifiers which use distinct pattern representations and show that many existing schemes can be considered as special cases of compound classification where all the pattern representations are used jointly to make a decision. An experimental comparison of various classifier combination schemes demonstrates that the combination rule developed under the most restrictive assumptions-the sum rule-outperforms other classifier combinations schemes. A sensitivity analysis of the various schemes to estimation errors is carried out to show that this finding can be justified theoretically.
5,670 citations
"Signature Verification Using Static..." refers methods in this paper
...Weighted sum rule is used as the multi-classifier decision algorithm to obtain the final matching result [5]....
[...]
Book•
02 Apr 2013
TL;DR: This book covers the general principles and ideas of designing biometric-based systems and their underlying tradeoffs, and the exploration of some of the numerous privacy and security implications of biometrics.
Abstract: Biometrics: Personal Identification in Networked Society is a comprehensive and accessible source of state-of-the-art information on all existing and emerging biometrics: the science of automatically identifying individuals based on their physiological or behavior characteristics. In particular, the book covers: *General principles and ideas of designing biometric-based systems and their underlying tradeoffs *Identification of important issues in the evaluation of biometrics-based systems *Integration of biometric cues, and the integration of biometrics with other existing technologies *Assessment of the capabilities and limitations of different biometrics *The comprehensive examination of biometric methods in commercial use and in research development *Exploration of some of the numerous privacy and security implications of biometrics. Also included are chapters on face and eye identification, speaker recognition, networking, and other timely technology-related issues. All chapters are written by leading internationally recognized experts from academia and industry. Biometrics: Personal Identification in Networked Society is an invaluable work for scientists, engineers, application developers, systems integrators, and others working in biometrics.
1,845 citations
TL;DR: Experiments on a database containing a total of 1232 signatures of 102 individuals show that writer-dependent thresholds yield better results than using a common threshold.
595 citations
01 Jan 1994
TL;DR: LTS1 Reference LTS-ARTICLE-1994-002 Record created on 2006-06-14, modified on 2016-08-08.
Abstract: Keywords: LTS1 Reference LTS-ARTICLE-1994-002 Record created on 2006-06-14, modified on 2016-08-08
156 citations
"Signature Verification Using Static..." refers methods in this paper
...The static behavior of the signature is analyzed using 1D - log Gabor wavelet transform and Euler numbers and matching is performed to obtain the matching scores for the offline features....
[...]
...The textural and topological features of a signature are extracted using algorithms based on 1D log Gabor and Euler numbers respectively....
[...]
...In the frequency domain, log-Gabor filter bank according to Bigun et al. [1] is defined as: ),(),( αα ϕϕ ωωωωω iirrrij GG −= (1) where ),( ϕr are polar coordinates, αω ir is the logarithm of the center frequency at scale i, αϕω i is the j th orientation and ),( ϕωω rG is defined as: ) 2 exp() 2 exp( 2 2 2 2 , jri r r G ϕ ϕ ωω σ ω σ ω ϕ = (2) where 2riσ and 2jϕσ are parameters of the Gaussian function....
[...]
...[1] is defined as: ) , ( ) , ( α α φ φ ω ω ω ω ω i i r r r ij G G − = (1)...
[...]
...1D-log Gabor [1] is used to extract the textural features of the signature pattern and Euler number is used to extract the topological features to compute the matching score for the static features....
[...]
TL;DR: Complex moments of the Gabor power spectrum yield estimates of the N-folded symmetry of the local image content at different frequency scales, that is, they allow to detect linear, rectangular, hexagonal/triangular, and so on, structures with very fine to very coarse resolutions as discussed by the authors.
Abstract: Complex moments of the Gabor power spectrum yield estimates of the N-folded symmetry of the local image content at different frequency scales, that is, they allow to detect linear, rectangular, hexagonal/triangular, and so on, structures with very fine to very coarse resolutions. Results from experiments on the unsupervised segmentation of real textures indicate their importance for image processing applications. Real geometric moments computed in Gabor space also provide for very powerful texture features, but lack the clear geometrical interpretation of complex moments. >
146 citations