1. Presented By: Dr. Hlaing Htake Khaung Tin Computer University (Loikaw) Myanmar 30 th November 2014 1 Biometric Research: Face Recognition and Age Prediction International Conference on Multidisciplinary Research & Practice (ICMRP-2014)

2. Outlines 2  Introduction  Objectives  Definition of the Problem  Comparison of Age Prediction Techniques  Real-World Applications  Steps in Age Prediction  Experimental Results  Conclusion

3. Introduction 3  Nowadays, as increasing security is being demanded in everywhere of the human society, so face recognition has become a very active in human biometrics research area.  Some examples of biometric features of human are: 1. Signature- studies the pattern, speed, acceleration and pressure of the pen when writing ones signature. 2. Fingerprint- studies the pattern of ridges and furrows on the surface of the fingertip. 3. Voice- studies way humans generate sound from vocal tracts, mouth, nasal cavaties and lips.

4. Introduction (Cont’d) 4 4. Iris- studies the annular region of the eye bounded by the pupil and the sclera. 5. Retina- studies the pattern formed by veins beneath the retinal surface in an eye. 6. Hand Geometry- measures the measurements of the human hand. 7. Ear Geometry- measures the measurements of the human ear. 8. Facial Thermogram- concerns the heat that passes through facial tissue. Among them face is the most natural and well know biometric.

5. 5  to create the fast recognition system for the face database  to predict how old the person is and to carry out the face recognition system based on this predicted age  to stop underage drinkers from entering bars, prevent minors from purchasing tobacco products from vending machines  children access to adult Web sites by predicting their age.  The main advantage of this system is reduction of searching time and it requires small memory usage.

6. Definition of the Problem 6  Implementing algorithms that enable the prediction of a person’s age; based on features derived from his/her face image. Age/Age Groups Table 1: Age Groups and Age Intervals

7. Comparison of Age Prediction Techniques 7 ResearcherDatabaseTrainingTestingAge Estimation Accuracy Kwon and Lobo47 images, classified by babies, young adults and senior adults -15 images100% Horng230 images, classified by babies, young adults 115 images 81.6% Lanitis330 images ranging from 0 to 35 years old 250 images80 images55% Yan-Ming Tanng1072 images ranging from =50 804 images268 images81.72% Proposed System 60 330 images220 images88% Table 2: Summary of Age Prediction

8. Related work 8 1. Y.H. Kwon and N.da Vitoria Lobo, “Age Classification from facial images”, Computer Vision and Image Understanding, Vol: 74, Page(s): 1-21, April 1999. 2. Wen-Bing Horng, Cheng-Ping Lee and Chun-Wen Chen, “Classification of Age Groups Based on Facial Features”, Journal of Science and Engineering, Vol: 4, Page(s): 183-192, 2001. 3. Hayashi, M. Yasumoto, H.Ito, Y.Niwa, and H.Koshimizu, “Age and Gender Estimation from Facial Image Processing”, the 41 st SICE Annual Conference, Vol:1, Page(s): 1-13, August 2002. 4. A.Lanitis, C.Draganova, and C.Christodoulou, “Comparing different classifiers for automatic age estimation”, IEEE TSMCB, Vol: 34, Page(s): 621-628, 2004. 5. Yan-Ming Tang and Bao-Lang Lu, “Age Classification Combing Contour and Texture Feature”, 2010.

9. Real-World Applications 9  There are many popular real-world applications related to age prediction. ForensicGovernmentCommercial Corpse Identification, Criminal Investigation, Terrorist Identification, Parenthood Determination, Missing Children, etc. National ID Card, Correctional Facility, Driver’s License, Social Security, Welfare Disbursement, Border Control, Passport Control, etc. Computer Network Logon, Electronic Data Security, E-Commerce, Internet Access, ATM, Credit Card, Cellular Phones, Personal Digital Assistant, Medical Records, Management, Distance Learning, etc. Table 3: Real-World Applications

10. 10  Age-specific access control o Vending machines that prevent minors from buying alcohol or cigarettes Figure 1: Vending machines

11. Steps in Age Prediction 11 Image Acquisition Preprocessing Face Feature Extraction Age Prediction Age Prediction with related age group Output Figure 2: Steps in Age Prediction System  Preprocessing is a very important factor in the overall face recognition system.  The preprocessing step reduces the irrelevant information, eliminates noise and analyses the input image. (a)Input Face Image(b)Face Region after Cropping Figure 3: Cropping the Face Area from the Input Face Image

12. Table 4. Database Specifications Database Specifications 12

13. Figure 4: Some Face Images in the training Database Experimental Results 13

14. Experiment 1 (For Person in Training Database) Figure 5: Experiment 1 14

15. Experiment 2 (For Person in Training Database) Figure 6: Experiment 2 15

16. Accuracy Graph of Age Prediction 16 Figure 7: The accuracy graph of age prediction results The accuracy rate is about 88%. The accuracy of the system can be analyzed by the variation on the range of the age groups. The larger errors occur at age 21 to 25 aging group and age 41 to 45 aging group.

17. Conclusion 17  The age dependent face recognition system is developed based on 11 individual aging classes, which yields a great reduction time complexity in search space than searching the entire database.  The algorithms that have been developed are tested on AT&T, Yale, MORPH, FG-NET Face Databases and other face database images which are collected from UCSY and the Internet.  The range of five-year age groups were used for age prediction.  The main advantage of this system is reduction of searching time and it requires small memory usage.  According to the experiment result, this system is an effective age dependent face recognition system.

18. Author’s Publications 18  [1]Hlaing Htake Khaung Tin, “Facial Extraction and Lip Tracking Using Facial Point”, International Journal of Computer Science and Engineering and Information Technology (IJCSEIT), vol.1, no.1, April 2011. (India) http://airccse.org/journal/ijcseit/current.html  [2]Hlaing Htake Khaung Tin and Myint Myint Sein, “Fast Facial Recognition System”, IEEE The proceeding of the 8 th International Joint Conference on Computer Science and Software Engineering (JCSSE 2011), May 2011. (Thailand)  [3]Hlaing Htake Khaung Tin and Myint Myint Sein, “Effective Method of Age Dependent Face Recognition”, The proceeding of International Conference on Computer Science and Informatics (ICCSI 2011), June 2011. (India)

19. Author’s Publications 19  [4]Hlaing Htake Khaung Tin and Myint Myint Sein, “Effective Method of Age Dependent Face Recognition”, International Journal of Computer and Informatics (IJCSI), vol.1, issue.1, June 2011. (India) http://interscience.in/ijcsi_Vol1Iss2.html  [5]Hlaing Htake Khaung Tin and Myint Myint Sein, “Race Identification from Face Images”, The proceeding of International Conference on Advances in Computer Engineering (ACE 2011), August 2011. (India)  [6]Hlaing Htake Khaung Tin and Myint Myint Sein, “Race Identification from Face Images”, International Journal on Information Technology (IJIT), vol.1, issue.2, August 2011. (India) http://connection.ebscohost.com/c/articles/82513488/race-identification-face- images

20. Author’s Publications 20  [7]Hlaing Htake Khaung Tin and Myint Myint Sein, “Robust Method of Age Dependent Face Recognition”, IEEE the 4 th International Conference on Intelligent Networks and Intelligent Systems (ICINIS 2011), November 2011. (China) (Best Paper Award Achievement)Best Paper Award Achievement  [8]Hlaing Htake Khaung Tin and Myint Myint Sein, “Aging Groups Classification based on Facial Feature”, The proceeding of International Conference on Information Technology Management Issues (ICITMI 2011), September 2011. (Malaysia)  [9]Hlaing Htake Khaung Tin and Myint Myint Sein, “Developing the Age Dependent Face Recognition System”, International Journal of Intelligent Engineering and Systems (IJIES), vol.4, issue.4, December 2011. (China) http://www.inass.org/ContentsPapers.html

21. Author’s Publications 21  [10]Hlaing Htake Khaung Tin and Myint Myint Sein, “Aging Group Classification based on Facial Feature”, Journal of Information Systems : New Paradigms (JISNP), vol.1, no.1, September 2011. (Malaysia) http://www.seaofsci.com/jisnp/currentissue.html  [11]Hlaing Htake Khaung Tin, “Perceived Gender Classification from Face Images”, International Journal of Modern Education and Computer Science (IJMECS), vol.4, no.1, February 2012. (Hong Kong) http://www.mecs-press.org/ijmecs/v4n1.html  [12]Hlaing Htake Khaung Tin, “Robust Algorithm for Face Detection in Color Images”, International Journal of Modern Education and Computer Science (IJMECS), vol.4, no.2, March 2012. (Hong Kong) http://www.mecs-press.org/ijmecs/v4n2.html

22. Author’s Publications 22  [13]Hlaing Htake Khaung Tin, “Subjective Age Prediction of Face Images Using PCA”, International Journal of Information and Electronics Engineering (IJIEE), vol.2, no.3, May 2012. (Singapore) http://www.ijiee.org/vol2no3.htm  [14]Hlaing Htake Khaung Tin, “Gender and Age Estimation Based on Facial Images”, International Journal : ACTA TECJNICA NAPOCENSIS Electronics and Telecommunications, vol. 52, no.3, 2011. (Romania) http://users.utcluj.ro/~atn/Numbers.htm

23. Author’s Publications 23  [15]Hlaing Htake Khaung Tin, “Feature based Age Prediction for Face Recognition”, The 5 th Parallel and Soft Computing (PSC 2010), December 2010. (UCSY)  [16]Hlaing Htake Khaung Tin and Myint Myint Sein, “Automatic Aging Simulation of the Human Face”, The 9 th International Computer Applications (ICCA 2011), May 2011. (Myanmar)  [17]Hlaing Htake Khaung Tin, “Removal of Noise Reduction for Image Processing”, The 6 th Parallel and Soft Computing (PSC 2011), December 2011. (UCSY)  [18]Hlaing Htake Khaung Tin, “Automatic Age Prediction of Aging Effects on Face Images”, The 10 th International Computer Applications (ICCA 2012), February 2012. (Myanmar)

24. Author’s Publications 24  [19]Hlaing Htake Khaung Tin, “Personal Identification and Verification using Palm Print Biometric”, The 3 rd International Conference on Advancement of Engineering (ICAE 2012), December 2012. (India)  [20]Hlaing Htake Khaung Tin, “Personal Identification and Verification using Palm Print Biometric”, International Journal of Latest Technology in Engineering Management and Applied Science (IJLTEMAS), vol. 1, issue. X, December 2012. (India) http://www.ijltemas.in/digital-library/index

25. Author’s Publications 25  [21]Hlaing Htake Khaung Tin, “Effective Method of Face Recognition and Palmprint Biometrics for Personal Identification”, International Journal of Advanced and Innovative Research (IJAIR), December 2012. (India)  [22]Hlaing Htake Khaung Tin, “Age Dependent Face Recognition using Eigenface”, International Journal of Modern Education and Computer Science(IJMECS), vol.5, no.9, October 2013. (Hong Kong)  [23] Hlaing Htake Khaung Tin and Myint Myint Sein, “How Old Are You?: Age Prediction using Eigen Face”, International Conference on Science and Engineering (ICSE2013), December 2013. (Myanmar)

26. Author’s Publications 26  [24] Hlaing Htake Khaung Tin and Myint Myint Sein, “Real time age prediction and gender classification”, The First International Conference on Energy Environment and Human Engineering (ICEEHE2013), December 2013. (Myanmar)  [25] Hlaing Htake Khaung Tin and Myint Myint Sein, “Automatic Race Identification from Face Images in Myanmar”, The First International Conference on Energy Environment and Human Engineering (ICEEHE2013), December 2013. (Myanmar)  [26] Hlaing Htake Khaung Tin, “What is your race?: Ethnicity and race identification in Myanmar”, International Conference on Research and Scientific Innovation (ICRSI2013), December 2013. (India).  [27] Hlaing Htake Khaung Tin, “Performance Evaluation for the Age Dependent Face Recognition System”, International Journal of Innovation and Scientific Research (IJISR), Volume 9, Issue 2, September 2014, Pages 481–486.

27. 27