signature recognition

File:Online signture.jpg

File:Dynamic information of a signature.jpg

Signature recognition is an example of behavioral biometrics that identifies a person based on their handwriting. It can be operated in two different ways:

Static: In this mode, users write their signature on paper, and after the writing is complete, it is digitized through an optical scanner or a camera to turn the signature image into bits.{{Cite journal|last1=Ismail|first1=M.A.|last2=Gad|first2=Samia|date=Oct 2000|title=Off-line arabic signature recognition and verification|url=http://dx.doi.org/10.1016/s0031-3203(99)00047-3|journal=Pattern Recognition|volume=33|issue=10|pages=1727–1740|doi=10.1016/s0031-3203(99)00047-3|bibcode=2000PatRe..33.1727I |issn=0031-3203}} The biometric system then recognizes the signature analyzing its shape. This group is also known as "off-line".{{Cite web|date=2016-01-11|title=Explainer: Signature Recognition {{!}} Biometric Update|url=https://www.biometricupdate.com/201601/explainer-signature-recognition|access-date=2021-04-03|website=www.biometricupdate.com|language=en-US}}

Dynamic: In this mode, users write their signature in a digitizing tablet, which acquires the signature in real time. Another possibility is the acquisition by means of stylus-operated PDAs. Some systems also operate on smart-phones or tablets with a capacitive screen, where users can sign using a finger or an appropriate pen. Dynamic recognition is also known as "on-line". Dynamic information usually consists of the following information:

• spatial coordinate x(t)
• spatial coordinate y(t)
• pressure p(t)
• azimuth az(t)
• inclination in(t)
• pen up/down

The state-of-the-art in signature recognition can be found in the last major international competition.{{cite journal|last=Houmani|first=Nesmaa |author2=A. Mayoue |author3=S. Garcia-Salicetti |author4=B. Dorizzi |author5=M.I. Khalil |author6=M. Mostafa |author7=H. Abbas |author8=Z.T. Kardkovàcs |author9=D. Muramatsu |author10=B. Yanikoglu |author11=A. Kholmatov |author12=M. Martinez-Diaz |author13=J. Fierrez |author14=J. Ortega-Garcia |author15=J. Roure Alcobé |author16=J. Fabregas |author17=M. Faundez-Zanuy |author18=J. M. Pascual-Gaspar |author19=V. Cardeñoso-Payo |author20=C. Vivaracho-Pascual |title=BioSecure signature evaluation campaign (BSEC'2009): Evaluating online signature algorithms depending on the quality of signatures|journal=Pattern Recognition|date=March 2012|volume=45|issue=3|pages=993–1003|doi=10.1016/j.patcog.2011.08.008|bibcode=2012PatRe..45..993H |s2cid=17863249 }}

The most popular pattern recognition techniques applied for signature recognition are dynamic time warping, hidden Markov models and vector quantization. Combinations of different techniques also exist.{{cite journal|last=Faundez-Zanuy|first=Marcos|title=On-line signature recognition based on VQ-DTW|journal=Pattern Recognition|year=2007|volume=40|issue=3|pages=981–992|doi=10.1016/j.patcog.2006.06.007|bibcode=2007PatRe..40..981F }}