Evaluation of Texture Descriptors for Validation of Counterfeit Documents

Centeno, Albert Berenguel; Terrades, Oriol Ramos; Canet, Josep Llados; Morales, Cristina Cañero

doi:10.1109/icdar.2017.204

Cited by 11 publications

(7 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The data set of all probabilities in HMM is given in Fig ( 9). Table (1) shows the path number (i), its formula, and its probability value where i=1, 2,….,16. Consequently, in Fig( 10) it was found that the most frequently occurring event is the most probable path 9, when the number of deposits is twice.…”

Section: Fig (9): All Probability Values Of All Paths For the States ...mentioning

confidence: 99%

“…That is to obtain the numerical results using the Markov model and the hidden Markov model to predict the values of banknotes and whether if they are fake or not. # sequence of observations >observations=c("notFake","Fake","notFake","notFake","notFake","Fake","Fake") >viterbi(hmm,observations) [1] "typeA" "typeA" "typeA" "typeA" "typeA" "typeA" "typeA"…”

Section: Appendixmentioning

confidence: 99%

“…This currency has experienced various developmental phases, culminating in its present form. Presently, banknotes have attained considerable significance, primarily due to the widespread confidence in their legitimacy and security, thereby facilitating their acceptance in transactions without apprehension regarding their authenticity or safety [1].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Prediction Using Markov Model and Hidden Markov Model for the States of Banknotes

Hendam,

Mahfouz,

Genedi

2024

Alfarama Journal of Basic & Applied Sciences

View full text Add to dashboard Cite

This study aims to solve the problem of identifying the values of banknotes, and whether they are fake or not. This problem is considered one of the topics that many researchers are interested in the field of detecting banknotes values and determining whether they are fake. For this reason, Markov models were applied to identify the probabilities of banknote values and whether they are fake or not. In addition, the hidden Markov model was used to identify the probabilities of banknotes states when they were deposited a limited number of times. Important numerical results were obtained in applying the Markov model and the hidden Markov model where it can be implemented in various other applications to be used in prediction. This study discovered the probabilities of banknote values and whether they are fake or not when depositing using the Markov models. In addition, the probabilities of tracing the trajectories of banknote states on deposit were obtained by the hidden Markov model. Researchers can apply this study to other data sets to predict the future data.

show abstract

Section: Fig (9): All Probability Values Of All Paths For the States ...mentioning

confidence: 99%

Section: Appendixmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Prediction Using Markov Model and Hidden Markov Model for the States of Banknotes

Hendam,

Mahfouz,

Genedi

2024

Alfarama Journal of Basic & Applied Sciences

View full text Add to dashboard Cite

show abstract

“…The authors used a lab-made dataset of forgeries by scanning genuine euro bills and then printed counterfeits with an HP LaserJet printer. In a follow-up paper [25], Berenguel et al proposed a different classification method, but still used the same procedure to generate lab-made counterfeit samples for evaluation. All of these papers have a common limitation that the results are only valid for the specific counterfeit samples used in the study.…”

Section: B Attack On Fingerprintsmentioning

confidence: 99%

“…First of all, we do not require any dataset of forgeries for training. Instead of merely classifying a banknote into a binary result of "real" or "forgery" like in [18]- [25], our technique extracts a unique fingerprint from a physical banknote. The authentication of a banknote starts with a null hypothesis that it is a "forgery" until this hypothesis is compellingly rejected by statistics.…”

Section: B Attack On Fingerprintsmentioning

confidence: 99%