Abstract Soft biometric attributes such asgender,ethnicityand agecan be determinedfrom theirisimages.Pupil size plays an important factor iniristemplate aging. In thisstudy,statisticalexperimentsareperformedtofind out confidence interval forIris-Pupilthickness ofdifferentage groupssuch as children, youthandsenior citizen.Significant group differences have been observed by applying statistical techniques such asAnalysis of Variance (ANOVA)and the Tukey€s pairwise comparison test.The results of the study conclude thatthe proposed methodology can beemployedto determineage group of a person from the available iris images.Basedonthe study results, we argue thatperformance ofaniris recognition system can be enhanced by identifying age groupof the personsfromtheiriris images.

References

[1]Abbasi A., Khan M., andKhan S., ACritical SurveyofIrisbasedRecognition Systems Middle-East Journal of Scientific Research,vol. 15, no.5,pp.663-668, 2013.

[2]Azizi A. andRezaH., Efficient IRIS Recognition through ImprovementofFeature Extraction Subset SelectionInternational Journal of Computer Science and Information Security,vol.2, no.1,pp. 1-10,2009.

[3]Belcher C.and Du Y., Region-based SIFT ApproachtoIris RecognitionOptics and Lasers in Engineering, vol. 47, no. 1, pp. 139-147, 2009.

[4]Czajka A., TemplateAgeinginIris Recognition,available at: http://zbum.ia.pw.edu.pl/PAPERS/BIOSIGNALS _2013_Czajka.pdf,last visited2013.

[5]FairhurstM.and Erbilek M., Analysis of Physical Ageing EffectsinIris Biometrics,IET Computer Vision,vol.5, no.6,pp.358-366, 2011.

[6]Fenker S. and Bowyer K., Analysis ofTemplate AginginIris Biometricsin ProceedingsIEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, Providence,pp. 45-51, 2012.

[7]FenkerS.andBowyer K., Experimental Evidenceof aTemplate Aging EffectinIris Biometrics,available at: https://www3.nd.edu/~kwb/FenkerBowyerCVPR W_2012.pdf,last visited2011.

[8]Fenker S., OrtizE.,andBowyer K.,available at: http://ieeexplore.ieee.org/stamp/stamp.jsp?arnum ber=6516567,last visited2013.

[9]He Y., Ma Z., and Zhang Y.,FeatureExtraction ofIrisbased onTexture Analysis,Advances in Future Computer and Control Systems,Springer, Berlin Heidelberg, 2012.

[10]Lagree S. andBowyer K., PredictingEthnicity andGenderfromIris Texturein Proceedings of IEEE International Conference on Technologies for Homeland Security,Waltham, pp. 440-445, 2011.

[11]Mahmud M., Khan M., Alghathbar K., Abdullah A., and Idris M., IntrinsicAuthenticationof Multimedia ObjectsusingBiometric Data ManipulationTheInternational Arab Journal of Information Technology, vol. 9, no. 4, pp. 336- 342, 2012.

[12]Ortiz E., Bowyer K.,and Flynn P., ALinear Regression Analysisof theEffectsofAge Related Pupil Dilation ChangeinIris Biometricsin Proceedings of the6th International Conference onBiometrics: Theory, Applications and Systems,Arlington,pp. 1-6, 2013.

[13]Sapthagirivasan V., Circular ROI Crop, availableat: http://www.mathworks.com/matlabcentral/fileexc hange/28867-circular-roi-crop,last visited2013. Iris-Pupil Thickness Based Method for Determining Age Group of a Person721

[14]Sgroi A., Bowyer K., and Flynn P., The PredictionofOldandYoung SubjectsfromIris Texturein Proceedings ofInternational Conference on Biometrics,Madrid,pp. 1-5, 2013.

[15]Ziauddin S. and Dailey M., ARobust Hybrid Iris Localization Techniquein Proceedings of the6thInternational Conference on Electrical Engineering/ Electronics,Computer, Telecommunications and Information Technology,Pattaya,pp. 1058-1061, 2009. Asima Abbasiobtained MS in Software Engineering from SZABIST, Islamabad. Her research interests are in the fields ofpattern recognition and software designand implementation. Muhammad Khanobtained DPhil degree inComputer System Engineeringfrom the University of Sussex, UK. His research interests include software engineering, cyber administration, information security policies, digitalforensic analysis and machine learning techniques.