A systematic comparison between on-line and off-line methods for signature verification with hidden Markov models

Rigoll, Gerhard; Kosmala, A.

doi:10.1109/icpr.1998.712066

Cited by 56 publications

(36 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…These results confirm on a public, replicable and objective benchmark, what was already pointed out in several previous works [8,18]: dynamic signature contains more information than its static version and, therefore, can lead to lower error rates when two highly competitive on-line and off-line recognition algorithms (systems A and C) are compared.…”

Section: Experiments 1 -Monomodal: Resultssupporting

confidence: 72%

“…The fusion of static and dynamic signature to enhance the performance of automatic recognition systems has already been studied in several works, where it has been shown that such a fusion approach can yield a significant decrease in the error rates [8,16,17,18]. Although all of them represent very valuable research efforts, in most of these previous approaches, experiments are carried out on small proprietary databases which do not contain real off-line data (static signatures are generated as single stroke images from the on-line version) or where on-line and off-line samples were not acquired simultaneously but on different sessions.…”

Section: Related Workmentioning

confidence: 99%

“…One of the first efforts that considered the combination of on-line and off-line features was conducted in [18]. The tests were carried out over a very limited database comprising 20 signatures per subject of 14 individuals.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

On-line signature recognition through the combination of real dynamic data and synthetically generated static data

Galbally

Diaz

Ferrer

et al. 2015

Pattern Recognition

107

View full text Add to dashboard Cite

This is the author’s version of a work that was accepted for publication in Pattern Recognition . Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Pattern Recognition , 48, 9 (2005) DOI: 10.1016/j.patcog.2015.03.019On-line signature verification still remains a challenging task within biometrics. Due to their behavioral nature (opposed to anatomic biometric traits), signatures present a notable variability even between successive realizations. This leads to higher error rates than other largely used modalities such as iris or fingerprints and is one of the main reasons for the relatively slow deployment of this technology. As a step towards the improvement of signature recognition accuracy, the present paper explores and evaluates a novel approach that takes advantage of the performance boost that can be reached through the fusion of on-line and off-line signatures. In order to exploit the complementarity of the two modalities, we propose a method for the generation of enhanced synthetic static samples from on-line data. Such synthetic off-line signatures are used on a new on-line signature recognition architecture based on the combination of both types of data: real on-line samples and artificial off-line signatures synthesized from the real data. The new on-line recognition approach is evaluated on a public benchmark containing both real versions (on-line and off-line) of the exact same signatures. Different findings and conclusions are drawn regarding the discriminative power of on-line and off-line signatures and of their potential combination both in the random and skilled impostors scenarios.M. D.-C. is supported by a PhD fellowship from the ULPGC and M.G.-B. is supported by a FPU fellowship from the Spanish MECD. This work has been partially supported by projects: MCINN TEC2012-38630- C04-02, Bio-Shield (TEC2012-34881) from Spanish MINECO, BEAT (FP7-SEC-284989) from EU, CECABANK and Cátedra UAM-Telefónic

show abstract

Section: Experiments 1 -Monomodal: Resultssupporting

confidence: 72%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

On-line signature recognition through the combination of real dynamic data and synthetically generated static data

Galbally

Diaz

Ferrer

et al. 2015

Pattern Recognition

107

View full text Add to dashboard Cite

show abstract

“…For the problems of signature verification, researchers are continuously introducing new ideas, concept, and algorithms in order to increase the accuracy up to 100%. A brief and systematic comparison between offline and online signature verification is compared based on Hidden Markov Models in [6], and [7]. Different methods for signature verification which extracts certain dynamic features derived from velocity and acceleration of pen together with global parameters like total time taken, number of pen ups and downs is proposed in [8], and [9].…”

Section: Previous Workmentioning

confidence: 99%

Offline Signature Verification for Authentication

Jana¹,

Mandal²,

Chhaya³

2015

IJCA

View full text Add to dashboard Cite

Biometrics verification has become a recent trend to prevent unauthorized accesses to all kinds of e-data. Signature is strongly accepted in legally and socially as identification and authentication of a person's identity. But, it is very difficult to verify the signature physically. So, it is needed to design a system that verifies the signature of a human automatically. A set of actual signatures is collected from individuals whose signatures have to be authenticated by the system. The topological and texture features are extracted from the actual signature set. The system is trained by using these features. The mean feature values of all the actual signature features are calculated. This mean features acts as the model for verification against a test signature. Euclidian distance between template signature features and claimed signature features serves as a measure of similarity between the two. If this distance is greater than a predefined threshold, then the test signature is detected as fake. The system provides the result to classify actual and forgery signature with accuracy up to 100%.

show abstract

“…Dynamic information is always available in case of online signature recognition, such as velocity, acceleration and pen pressure which is more difficult to estimate than the Static shape of signature [1]. For the offline signature recognition systems [2] [3] [4], the previously written signature are captured by scanning or by other biometric system as a static image and then the recognition is carried out.…”

Section: Introductionmentioning

confidence: 99%

Offline Signature Verification using FMCN with GA based Optimization of Features

Jain¹,

Patil²

2012

IJCA

View full text Add to dashboard Cite

In recent years, along with extraordinary diffusion of internet and growing need of personal identification in many applications, signature verification is considered with interest. This paper proposed an offline signature verification method based on Genetic Algorithm and Fuzzy Min Max Neural Network Classifier with Compensatory Neuron. The proposed method is basically consists of two steps. At first step optimizing the features using genetic algorithm, and at second step signature recognition is done using Fuzzy Min Max Neural Network Classifier with Compensatory Neurons. The sample of signatures is used to represent a particular person. The sample signature is first preprocessed, and then features of the processed signature are extracted by using Krawtchouk moment. After feature extraction, these features are optimized by using genetic algorithm and finally optimized features are given to the classification phase for recognition. With this proposed method, we get the 98% accuracy in recognition and less time is required for classification with optimized features as compared to time required for classification without optimizing feature.

show abstract

A systematic comparison between on-line and off-line methods for signature verification with hidden Markov models

Cited by 56 publications

References 3 publications

On-line signature recognition through the combination of real dynamic data and synthetically generated static data

On-line signature recognition through the combination of real dynamic data and synthetically generated static data

Offline Signature Verification for Authentication

Offline Signature Verification using FMCN with GA based Optimization of Features

Contact Info

Product

Resources

About