Master Thesis - Department of Computer Science

More documents

Recommendations

Info

improve the accuracy further. They obtained a maximum accuracy of 96.67% using ICA2 on a database (for expression variations) comprised of 272 images from CMU PIE and rest 328 from FERET. For finding subbands insensitive to illumination they experimented on a database consisting of 272 images from CMU PIE and remaining 60 images from Yale database and reported a maximum accuracy of 77.71%. The main drawback of this method lies in the adhoc method of selecting successful subbands on a testing set which involves over-tuning to attain maximum possible accuracy. None of these approaches have attempted to exploit the fact that, if some of the common visual features of a face exist in a certain subband and are suppressed during reconstruction, we obtain a subband face with only the discriminatory information. The retina of the human visual system has been observed to perform local multi-resolution, multi-channel processing of the signal using a bank of tuned band- pass filters [65, 85, 98]. The visual cortex uses second order relational differences in structure with respect to an average face for perception [41, 56]. Hence, a unified computational model consisting of DWT/IDWT (signal processing) and PCA/LDA (statistical processing) will be able to closely imitate the human visual system, more effectively than what statistical processing does alone. These studies motivated us to explore the use of IDWT to obtain a subband face, by suppressing an approximation and retaining some of the discriminating subbands in the process of image reconstruction, and then use the subband face for recognition using statistical approaches. In this chapter, we propose the subband face as a new representation for the face recognition task. Only the discriminatory information of a face is retained or captured in a subband face. Discrete wavelet transform is used to decompose the original face image into approximation and detail subbands. We perform multi-level dyadic decomposition of a face image using the Daubechies filters [30]. The subband face may be reconstructed from selected subbands by suppressing the approximation at a suitable higher level and retaining the details. An inherent information fusion is being performed by the reconstruction process which retains only the inter-class discriminatory informations and discards the inter-class common informations. The information of a face in the details at lower levels of decomposition, usually contains noise and redundant pixels which do not constitute any discriminatory information 46
for a face. This was illustrated by experimentation in [94] where the reconstruction of a face image from subbands upto a certain level, below that of the original face image at the root, provides better performance for face recognition. It is thus often necessary to eliminate the details for the representation of a face, and in such cases the subband face is obtained by partial reconstruction. In this chapter, we provide an algorithm to obtain optimum subject-specific subbands on the basis of minimization of four different criteria. Results of our experimentation presented in section 3.4 will validate a hypothesis that “subband face representation provides efficient face recognition”. Our approach is different from that suggested in [36, 40, 26, 140] where the individual subbands of the face image have been used as features separately or unified by subspace analysis techniques for face recognition. None of the past approaches deals with a method to reconstruct a subband face using IDWT (synthesis filter bank). Face image Subband faces LSA (Linear Subspace Analysis): Subbands DWT IDWT LSA PCA, LDA, 2D−PCA, 2D−LDA, DCV. (a) Features (b) Subband selection Find nearest neighbor Templates from training set Subband face Distance Figure 3.1: (a) Block diagram for generating subband face. The face image is decom- posed using Discrete Wavelet Transform (DWT) into subbands. Selected subbands are used to reconstruct the subband face using inverse DWT. (b) Subband faces are projected onto a lower dimension featurespace using PCA or LDA (or their variants) and matched using the nearest neighbor strategy. 47
Page 1 and 2:
DEVELOPMENT OF EFFICIENT METHODS FO
Page 3 and 4:
To My Parents, Sisters and Dearest
Page 5 and 6:
Mirnalinee, Shreyasee, Lalit, Manis
Page 7 and 8:
LDC Linear Discriminant Classifier
Page 9 and 10:
3 An Efficient Method of Face Recog
Page 11 and 12:
A.6 Minutiae Matching . . . . . . .
Page 13 and 14:
4.1 Effect of increasing number of
Page 15 and 16: List of Figures 2.1 Summary of appr
Page 17 and 18: 3.7 Area difference between genuine
Page 19 and 20: Abstract Biometrics is a rapidly ev
Page 21 and 22: CHAPTER 1 Introduction The issues a
Page 23 and 24: 1.1.1 Applications Biometrics has b
Page 25 and 26: • Spoof Attacks: An impostor may
Page 27 and 28: • A new approach for multimodal b
Page 29 and 30: CHAPTER 2 Literature Review This re
Page 31 and 32: ange [23], infrared scanned [137] a
Page 33 and 34: u 2 x 2 u 1 x 1 (a) PCA basis (b) P
Page 35 and 36: PCA Projection Class 1 Class 2 LDA
Page 37 and 38: F DIFS DFFS F (a) (b) F 1 L Figure
Page 39 and 40: image A of m rows and n columns is
Page 41 and 42: faces of 40 subjects. • Others: A
Page 43 and 44: malizing the vector components, the
Page 45 and 46: Figure 2.5: A fingerprint image wit
Page 47 and 48: features (gradient coherence, inten
Page 49 and 50: Figure 2.7: Examples of minutiae; A
Page 51 and 52: according to the local orientation
Page 53 and 54: • Parallel Mode: This operational
Page 55 and 56: can address the problem of noisy se
Page 57 and 58: Prior to Matching Sensor Level Feat
Page 59 and 60: determined by logistic regression.
Page 61 and 62: Class-indifferent Methods • Decis
Page 63 and 64: 3. The final DS soft vector is calc
Page 65: on three face databases and compare
Page 69 and 70: Image Rows h(.) g(.) 2 2 Columns Fi
Page 71 and 72: in the accuracy of face recognition
Page 73 and 74: (a) (a1) (a2) (a3) (b) (b1) (b2) (b
Page 75 and 76: Genuine Impostor Scores Scores 1 0
Page 77 and 78: Error d1 = (1 − TZeroF RR) d2 = |
Page 79 and 80: The algorithm for obtaining the sub
Page 81 and 82: each subject, 42 samples (flashes f
Page 83 and 84: Table 3.3: Peak Recognition Accurac
Page 85 and 86: Table 3.5: Peak Recognition Accurac
Page 87 and 88: Table 3.7: Performance of Ekenel’
Page 89 and 90: Table 3.9: Performance of Ekenel’
Page 91 and 92: to our proposed method. Section 4.2
Page 93 and 94: plored both of the spaces to captur
Page 95 and 96: project only the class means in ran
Page 97 and 98: ΩNull and ΩRange represents the
Page 99 and 100: Thus, calculation of the QR factori
Page 101 and 102: 4.3.3 Algorithm for Feature Fusion
Page 103 and 104: The architecture of our proposed me
Page 105 and 106: T R T S denoted by RVNull and RVNul
Page 107 and 108: x DNull DRange DP (x) DNull(x) DRan
Page 109 and 110: C 2 C 1 C 1 C 2 C X X 2 M X M X 2 1
Page 111 and 112: iii) Project all class means onto n
Page 113 and 114: Table 4.1: Effect of increasing num
Page 115 and 116: Table 4.3: Effect of increasing num
Page 117 and 118:
(Original face) (Null face) (Range
Page 119 and 120:
Table 4.6: Performance of Dual Spac
Page 121 and 122:
5.1 Introduction In recent years, t
Page 123 and 124:
This means for a classifier D: r i
Page 125 and 126:
validation1, validation2 and test.
Page 127 and 128:
5.3.1 The Algorithm The steps of ou
Page 129 and 130:
5.4 Experimental Results In this ch
Page 131 and 132:
sets. • We put as much as possibl
Page 133 and 134:
argument is invalid in the context
Page 135 and 136:
LDA and LDA produce the same accura
Page 137 and 138:
space. If the training data uniform
Page 139 and 140:
face for each subject, and compare
Page 141 and 142:
also be introduced. But class speci
Page 143 and 144:
These stages are discussed in the f
Page 145 and 146:
(a) (b) Figure A.2: Input fingerpri
Page 147 and 148:
(a) (b) Figure A.4: Orientation ima
Page 149 and 150:
(i) For each block centered at (i,
Page 151 and 152:
(a) (b) Figure A.7: Enhanced images
Page 153 and 154:
(a) (b) Figure A.10: (a) and (b) sh
Page 155 and 156:
(a) (b) Figure A.12: Binarized imag
Page 157 and 158:
(a) (b) Figure A.14: Thinned images
Page 159 and 160:
(a) (b) Figure A.16: Thinned finger
Page 161 and 162:
Paired Minutiae Minutiae with unmat
Page 163 and 164:
Bibliography [1] FVC 2002 website.
Page 165 and 166:
[25] Li-Fen Chen, Hong-Yuan Mark Li
Page 167 and 168:
[49] Lin Hong, Yifei Wan, and Anil
Page 169 and 170:
[71] L. Lam and C. Y. Suen. Applica
Page 171 and 172:
IEEE Tran. on Pattern Analysis and
Page 173 and 174:
of-the-Art Systems. IEEE Tran. on P
Page 175 and 176:
and Human Communication, pages 374-
show all

Master Thesis - Department of Computer Science

Create successful ePaper yourself

Delete template?

Save as template?