International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004 
  
System accuracy vs. No. of eigenfaces 
9 
     
1100 100 
Se 
> 90-7 
5 
g 
2 80 - 
o 
o 
< 70 
5 
= 60 
e 
50 
-1 1 3 5 7 9 
Number of eigenfaces in database 
Figure 5: Accuracy base on Eigenfaces 
System accuracy vs. No. of face 
orientaions 
89 + 
88.5 | 
88 
87.5 | 
Accuracy 
© 
co 9 o 
O C! 
855 | 
© 
C 
84.5 
0 5 10 15 
Number of face orintations 
Figure 6 System Accuracy vs Number of Face orientations 
6. CONCLUSION 
Face recognition has been an attractive field of research for 
engineering, computer vision scientists and security purposes.. 
Humans are able to identify reliably a large number of faces 
and scientists are interested in understanding the perceptual 
and cognitive mechanisms at the base of the face recognition 
process. Since 1888, many algorithms have been proposed as 
a solution to automatic face recognition. Although none of 
them could reach the human recognition performance. This paper 
presented an algorithm for face recognition by performing PCA on 
Wavelet Transform. The Wavelet Transform is used to decompose 
the original image into four Wavelet subbands, each with a 
different frequency component. PCA is then applied on this 
Wavelet to reconstruct the image into vector representation. 
Wavelet Transform provides an excellent image decomposition 
and texture description. The combination of Wavelet Transform 
and PCA gives a better recognition accuracy and significant 
performance improvement when the database has large number of 
images. It reduces computational load and increases accuracy of 
the system. The paper has resulted in an overall success being able 
783 
to perform reliable recognition in a constrained environment. a 
recognition accuracy of 86% has been achieved. While the 
problem of recognizing faces under gross variations remains 
largely unsolved, a thorough analysis of the strengths and 
weaknesses of face recognition using PCA has been presented and 
discussed. 
REFERENCES 
Zhang, D. 2000. Zhang. Automated Biometrics: Technologies and 
Systems, Kluwer Academic Publishers, pp. 137-142. 
Chen, C. F, Tsent, Y. S. and Chen Y. C., 2003. Combination of PCA 
and Wavelet Transform for Face Recognition on 2.5D Images. 
Image and Vision Computing NZ. Palmerston North, pp. 343-347. 
Gonzalez, R. C. and Woods, R. E. 2003. Digital Image 
Processing, 2" Edition, Addison-Wesley Publishing Company. 
Daubechies, I., 1992. Ten Lectures on Wavelets, Society for 
Industrial and Applied Mathematics, Philadelphia. 
Goswami, J. C. Chan A. K. 1999. Fundamentals of Wavelets, 
Theory, Algorithm and Applications, John Wiley and Sons, pp.. 
210-266. 
Rioul, O. and Vetterli, M. 1991. Wavelet and signal 
Processing, /EEE SP Magazine, pp. 14 — 37. 
Rioul, O. and Duhamel, P., 1992. Fast Algorithms for 
Discrete and Continuous Wavelet Trnasforms, /EEE Trans. 
On Information Theory, Vol. 38, No. 2, pp. 569-586. 
Mallat, S. G., 1989. A Theory for Multiresolution Signal 
decomposition: The wavelet representation, IEEE Trans. 
On Pattern Analysis and Machine Intelligence, Vol. 11, 
NO. 7, pp. 674 — 693. 
Mallat, S: G., 1989. Multifrequency Channel 
Decompositions of Images and Wavelet Models, /EEE 
Trans. Acoust., Speech, Signal Processing, Vol. 37, No. 12, 
pp. 2091-2110. 
Christophe G., Giorgos Z. and Giorgos T. A Wavelet Based 
Framework for Face Recognition, Foundation for Research and 
Technology-Hellas (ICS), Crete, Greece. 
Zhao, W., Chellappa, R, Rosenfeld, A. and Phillips, P.J. 2000 
Face Recognition: A Literature Survey, UMD CFAR Technical 
Report CAR-TR-948. 
Singh, R. K. and Rajagopalan, A. N. 2002. Background Learning 
For Robust Face Recognition, 76th International Conference on 
Pattern Recognition (ICPR). August 1-15. 
Hong, L 1998 Integrating faces and fingerprints for personal 
identification, IEEE Trans. PAMI, Vol 20, pp. 1295-1307.