Real time face detection and recognition

Paper conference

1. “ Real Time Face Detection And Recognition ” Guided by Prof. V. G. Raut A Presentation on By Mr. Pankaj R. Bhusari DEPARTMENT OF ELECTRONICS AND TELE-COMMUNICATION SINHGAD COLLEGE OF ENGINEERING PUNE October 2013

2. OUTLINE which biometric is the best? objectives introduction research challenges literature survey proposed work edge contour preprocessing for face detection performance evaluation metric different database flowchart application CONCLUSION REFERENCES

3. Which biometric is the best ? Face Detection: It is non intrusive process Cheap to implement Reliable DNA, Fingerprint, Iris, Signature, Voice Detections: It is intrusive process Costly for implementation Not reliable Vs

4. objectives To design a fully automatic system. The system should work in real time. System should have very high accurate face detection rate and low false recognition rate. System should work on large database.

5. introduction Difference between face detection and face recognition: Detection ( Two class classification) Face versus Non-face from an image Recognition ( Multi class classification) One person versus all the others from the database

6. Research challenges Facial Expression Image Orientation Imaging Condition Occlusion Pose Presence of Structural Components

7. Literature survey Year Of Publishing Title Preprocessing Feature Extraction Advantages Disadvantages Guoshing Huang, Jiahong Su (2008) A Real Time Face Detection And Tracking Skin colour analysis Head tracking Elliptical head tracking More feasible Circuitry required is complex K.T. Talele , Sunil Kadam (2009) Face Detection And Geometric Face Normalization Face detection and normalization Geometric normalization Real time system Computations are very difficult Rong Bao chen , ShiJie Zhang ( 2010) Video Based Face Recognition Technology For Auto Motive Security Face recognition Adaboost 2DFLD Nearest neighbor rule System is totally non-intrusive It is slow process

8. Proposed work Mobile Or Webcam OUTPUT Mr. Pankaj Prof. Bhusari Face Database Face Detection Face recognition

9. Edge contour U se for object detection. Here we are using edge contour to find or detect body from the capture image. To avoid face like structure from image or video so that we increases false acceptance rate and to reduce false rejection. Edge contour use histogram of oriented gradients (HOG) Algorithm.

10. Haar like features for face detection Edge Feature Line Feature Center-surround Feature

11. Adaboost algorithm for face detection Training Set (sub-windows) Integral Representation Feature computation AdaBoost Feature Selection Cascade trainer Strong Classifier 1 (cascade stage 1) Strong Classifier N (cascade stage N) Classifier cascade framework Strong Classifier 2 (cascade stage 2) FACE IDENTIFIED

12. Preprocessing for face detection No Preprocessing (Original image) RGB to grayscale image Histogram Equalization (HE) Histogram Equalization and Laplacian of Gaussian filter (HE+LG) Histogram Equalization and Laplacian of Gaussian filter and Contrast Adjustment (HE+LG+CA)

13. COMPLETE PRE PROCESSING STEP

14. Principal component analysis (PCA) “PCA is a mathematical tool for finding directions in which a distribution is stretched out.” Widely used in practice Gives best-known results

15. Performance evaluation metric Probe face: Test face images whose at least one other image is present in training dataset. Imposter face: Test face image whose no image is present in training dataset. False reject rate (FRR): FRR denotes the probability that a biometric system will not identify an identity of an individual or will fail to accept a probe face. False acceptance rate (FAR): FAR denote the probability that a biometric system will incorrectly identify an individual or will fail to reject an imposter. True positive (TP): It is also called as hit or a correctly detected face. False positive (FP): It is also called miss or false-detection, detecting a face where there is no face actually. False negative (FN): When missing a visible face. True negative (TN): Describing non-face regions correctly as non-face region

16. Different database Data set Location Description MIT Database (163) ftp://whitechapelrious.media.mit.edu/pub/images/ Faces Of 16 People, 27 of Each Person Under Various Illumination Conditions, Scale And Head Orientation. Feret Database (115) http ://www.nist.gov/humanid/feret A Large Collection of Male And Female Faces. Each Image Contains A Single Person With Certain Expression Umiste Database (56) http://images.ee.umist.ac.uk/danny/database.html 564 Images of 20 Subjects. Each Subject Covers A Range of Poses From Profile To Frontal Views. University of Bern Database ftp://iamftp/unibe.ch/pub/images/faceimages 300 Frontal Face Images People(10 Images Per Person) And 150 Profile Face Images (5 Images Per Person) Yale e Database (7) http://http://cvc.yale.edu Face Images With Expression, Glasses Under Different Illuminations Conditions. At&T (Olivetti) Database (136) http://uk.research.att.com 40 Subjects,10 Images Per Subject Harvard Database (57) ftp://ftp.hrl.harvard.edu/pub/faces Cropped, Masked Face Images Under A Wide Range Of Lightning Conditions M2VTS Database (116) http://poseidon.csd.auth.gr/M2VTS/index.html A Multimodal Database Containing Various Image Sequences.

17. Flowchart Images or video frame Edge Contour Face Detection Preprocessing Stage Feature Extraction Face Recognition Non Face Face Implement on Raspberry Pi Module DISPLAY

18. Application Government Use Law Enforcement Counter Terrorism Immigration Legislature Human Robot Interaction Information Security Commercial Use Day Care Gaming Industry Residential Security E-Commerce Voter Verification Banking Surveillance

19. Conclusion In this way we have studied preprocessing on captured image with different preprocessing techniques such as histogram of equalization, laplacian of gaussian filter, contrast adjustment etc. In the last we have got the processed image which will be use for feature extraction purpose in order to find and recognize face from database.

20. References Zhang, H. Peng , J. Zhou, and S. K. Pal, “A Novel Face Recognition System Using Hybrid Neural and Dual Eigen-spaces Methods ,” IEEE Trans Systems, Man, and Cybernetics, Part A , vol. 32, no. 6, pp. 787-793, Nov. 2002. G. Shakhnarovich and B. Moghaddam , “Face recognition in subspaces,” in Handbook of Face Recognition, Springer, New York, NY, USA, 2004. Jens Fagertun (2005) Face Recognition Master Thesis IMM. Peter M. Corcoran and Claudia Iancu (2011), “Automatic Face Recognition System for Hidden Markov Model Techniques”, New Approaches to Characterization and Recognition of Faces, Dr. Peter Corcoran (Ed.), ISBN: 978-953-307-515-0. Zhen lei, chao wang , qinghai wang , “Real Time Face Detection and Recognition for Surveillance Applications”, vol. 28, pp.142-146, no. 2009. C. Mei, E. Sommerlade , G. Sibley, P. Newman, and I. Reid, “Hidden View Synthesis using Real-Time Visual SLAM for Simplifying Video Surveillance Analysis,” in IEEE International Conference on Robotics and Automation , May 2011, pp.4240–4245. RongBao Chen and ShiJie Zhang, “Video-based Face Recognition Technology for Automotive Security”, vol. 40, no. 2, 2010. J. Ruiz-del-Solar and P. Navarrete , “Eigen-space-based Face Recognition: a Comparative Study of Different Approaches,” IEEE Trans. Systems, Man, and Cybernetics, Part C, vol. 35, no. 3, pp.315-325, Aug. 2005. P. Kakumanu , S. Makrogiannis , and N. Bourbakis , “A Survey of Skin-Color Modeling and Detection Methods,” Pattern Recognition, vol. 40, no. 3, pp.1106-1122, March 2007. Zhen lei, chao wang , qinghai wang , “Real Time Face Detection and Recognition for Surveillance Applications” , vol. 28, pp.142-146, no. 2009.

21. P. Kakumanu , S. Makrogiannis , and N. Bourbakis , “ A Survey of Skincolor Modeling and Detection Methods, ” Pattern Recognit ., vol. 40, no. 3, March 2007 , pp.1106-1122. P. Shihand and C. Liu, “ Face Detection Using Discriminating Feature Analysis and Support Vector Machine, ” Pattern Recognit ., vol. 39, no. 2, Feb 2006, pp.260-276. M. Kirby and L. Sirovich , Application of the KL procedure for the characterization of human faces, IEEE Trans. Pattern Anal. Machine Intell . 12 (1990) (1), pp. 103 – 108 K. Tan and S. Chen, Adaptively weighted sub-pattern PCA for face recognition, Neurocomputing 64 (2005), pp. 505 – 511 Lay, David. (2000). Linear Algebra and its Applications. Addison-Wesley, New York p. 441-486) R. Lienhart and J. Maydt , “ An Extended Set of Haar-like Features for Rapid Object Detection ” , IEEE International Conference on Image Processing, 2002 . M. Pham, Y. Gao , V. Hoang, and T. Cham, “ Fast polygonal integration and its application in extending haar -like features to improve object detection, ” in IEEE Conference on Computer Vision and Pattern Recognition . IEEE , 2010, pp. 942 – 949. N. Dalal and B. Triggs , “ Histogram of Oriented Gradients for Human Detection ” , in Proc. IEEE Computer Vision and Pattern Recognition (CVPR) , June 2005, vol. 1, pp. 886-893. R. C. Gonzalez and R. E.Woods , Digital Image Processing, Third Edition, 2008. PEM P. J. Phillips, P. Rauss , S. Der , FERET (Face Recognition Technology) Recognition Algorithm Development and Test Report, Technical Report ARL-TR 99 5, U.S. Army Research Laboratory. A. C. Loui , C. N. Judice , S. Liu, An Image Database for Benchmarking of Automatic Face Detection and Recognition Algorithms, Proc. IEEE Conference on Image Processing, Vol. 1, pp.146-150, 1998. http://www.biometricsinstitute.org http://www.face-rec.org http://www.icao.int/mrtd/overview/overview.cfm . MachineReadableTravel Documents (MRTD). References

22. Thank you! Any question?