Professional Documents
Culture Documents
LZM Face Recog
LZM Face Recog
Evangelos Sarıyanidia , Volkan Dağlıb , Salih Cihan Tekc , Birkan Tunçb , Muhittin Gökmenc
a
Department of Control Engineering, Istanbul Technical University, Istanbul. sariyanidi@itu.edu.tr
b
Informatics Institute, Istanbul Technical University, Istanbul. {tuncbi, daglivo}@itu.edu.tr
c
Department of Computer Engineering, Istanbul Technical University, Istanbul. {tek, gokmen}@itu.edu.tr
ABSTRACT In [8], Pseudo ZM invariants are used to extract feature vectors from
faces. Another ZM based approach that considers spatial locality by
In this paper, we propose a new image representation called Local partitioning input images is presented in [9]. This approach involves
Zernike Moments (LZM) for face recognition. In recent years, lo- dividing the face images into non-overlapping subregions and then
cal image representations such as Gabor and Local Binary Patterns extracting features from each of these subregions using pseudo ZMs.
(LBP) have attracted great interest due to their success in handling In this paper, we present a novel representation based on local
difficulties of face recognition. In this study, we aim to develop an al- Zernike moments (LZMs) and demonstrate the superiority of this
ternative representation to further improve the face recognition per- representation for face recognition. The moments are calculated at
formance. We achieve this by utilizing Zernike Moments which have every pixel by considering its local neighborhood. A complex mo-
been successfully used as shape descriptors for character recogni- ment image of equal size as the input image is obtained for each mo-
tion. We modify global Zernike moments to obtain a local represen- ment component resulting in a set of images, as shown in Figure 1.
tation by computing the moments at every pixel of a face image by Then, each moment image is divided to non-overlapping subregions
considering its local neighborhood, thus decomposing the image into and phase-magnitude histograms are extracted from the complex
a set of images, moment components, to capture the micro structure output of each moment at each subregion. The final face represen-
around each pixel. Our experiments on FERET face database reveal tation is obtained by concatenating the extracted phase-magnitude
the superior performance of LZM over Gabor and LBP representa- histograms.
tions.
Index Terms— Face recognition, feature extraction, Zernike 2. GLOBAL ZERNIKE MOMENTS
moments, regional histograms
ZMs of an image are defined as the projection of the image onto
1. INTRODUCTION a orthogonal set of polynomials called Zernike polynomials. The
Zernike polynomials are defined as
Robust 2D face recognition is one of the most active research ar- Vnm (ρ, θ) = Rnm (ρ)ejmθ (1)
eas in computer vision. Despite the impressive advances in recent
years, it still remains as a challenging problem, mainly because of where n is the order of the polynomial and m is the number of itera-
the amount of possible variations in the appearance of a face that tions such that |m| < n and n − |m| is even. The radial polynomials
can result from illumination conditions, viewpoint and facial expres- Rnm are given as
sions. Robust face recognition requires robust features with the abil-
n−|m|
ity to deal with such variations. Most face recognition approaches (−1)s ρn−2s (n − s)!
2
adopt feature extraction methods involving local shape descriptors Rnm (ρ) = . (2)
like Gabor filtering-based features[1, 2] and LBP[3] to handle some s=0 s!( n+|m|
2
− s)!( n−|m|
2
− s)!
of these variations. In this study, we aim to develop a simple yet
powerful representation as an alternative to these representations by The ZMs of a digital image f (i, j) are calculated as
utilizing moments.
n+1
N −1 N −1
Image moments are regarded as well established shape descrip- Znm = f (i, j)V (ρij , θij )Δxi Δyj , (3)
tors and have been frequently used for content based retrieval and π i=0 j=0
pattern recognition tasks. One of the most inspiring early studies
illustrating the potential of image moments is the study of Hu [4], where xi and yj are the image coordinates mapped to the range
y
in which the moment invariants that enable successful recognition [−1, 1], ρij = x2i + yi2 , θij = tan−1 xji and Δxi = Δyj =
against scaling, translation, and rotation are presented. √
2/N 2. The terms n+1 π
, Δxi , and Δyj are constant and they will
In [5], the usage of Zernike moments (ZMs) for character recog- be ignored in the remainder of this paper for clarity.
nition resulted in very promising recognition rates. In [6] and [7], The global form of ZMs expressed in (3) was used for face
the authors proposed methods to extract features from face images recognition in [7]. The authors utilized both the phase and magni-
through ZMs (ZMs) and discussed the rotation invariancy of ZMs. tude information. However the results reported on the FERET face
This work is partially supported by The Scientific and Technological
database [10] are not promising enough to demonstrate the real po-
Research Council of Turkey with the grant number 109E268. The third and tential of ZMs.
the fourth authors are partially supported by ITU-BAP with grants 34120 and To further investigate the potential of ZMs, we have divided
34385, respectively. the whole image into non-overlapping subregions and calculated the
Zkmn (i, j) = k
f (i − p, j − q)Vnm (p, q). (4) gathered from different moment components. This representation
has been shown to be very successful for face recognition [2, 3].
p,q=− k−1
2
Figure 3 illustrates the whole process.
This transformation provides a rich image representation by suc-
cessfully exposing the intensity variations around each pixel. The 4. EXPERIMENTAL RESULTS
components of the representation (i.e. complex images correspond-
ing to different moment orders) seem to be very robust to illumina- We have carried out extensive experiments to assess the performance
tion variations as shown in Figure 1. The complex and real parts of of our approach. The complex output of ZM images is represented
k
an exemplar Vnm filter for size k = 3, n = m = 1 are shown in using phase-magnitude histograms. The performance of the method
Figure 2. is substantially increased by applying the LZM transformation twice.
The number of components depends on the moment order n. We
impose an additional constraint on the second parameter m, such as
m = 0 since the imaginary part of the kernels Vnm k
becomes zero 4.1. Experimental Setup
when m = 0 and this is not a desirable behavior when the outcome In our method, there are several parameters that need to be de-
of (4) is used to extract the phase-magnitude histograms as explained termined, such as the size of ZM-based operator k, the moment
in Section 4.2. The number of effective moment components can be order n, the grid size N × N and the number of bins in the his-
calculated through the following expression: togram b. After extensive experiments, we found that good per-
n(n+2) formance is obtained when setting k = 7, n = 4, N = 10,
4
if n is even b = 24. When the moment order, n, is set to 4 the number of
K(n) = (n+1) 2 (5)
4
if n is odd. resulting moments, K, is 6, corresponding to moment components
Zk11 , Zk22 , Zk31 , Zk33 , Zk42 , Zk44 . The length of the final feature vector
is (N 2 + (N − 1)2 ) × b × K = 26064.
3.2. Face description from moment components Znm
All tests are performed on the FERET face database [10] using
The advantage of dividing images into subregions has been ex- standard FERET evaluation protocol: a gallery set of 1196 images
pressed in many face recognition studies. In this study, we follow of 1196 subjects and four probe sets: FaFb (1195 images), FaFc
586
Fig. 3. An illustration of the proposed method. Only magnitude images are shown. During histogram construction both magnitude and phase
values are considered.
(194 images), Dup1 (722 images) and Dup2 (234 images). The input transformed images. In other words, we apply the LZM transforma-
images are normalized to have zero mean and unit variance after they tion twice: First to stimulate the local shape characteristics; second,
are cropped, and their sizes are fixed to 130 × 150. The final feature to describe the local shape statistics of the transformed images. The
vectors are compared using L1 distance. moment components obtained by the cascaded LZM transformation
are shown in Figure 4, while the block diagram of the cascaded trans-
form is shown in Figure 5.
4.2. Local phase-magnitude histograms (H-LZM)
587
Fig. 5. An illustration of the cascaded approach. Magnitude images are shown after the second transformation. During histogram construction
both magnitude and phase values are considered.
rectly leads to better results and therefore report the results only phase-magnitude histograms of LZMs are very useful in obtaining
for these inputs in Table 2, where H-LZM2 -I, H-LZM2 -R, and H- successful face recognition schemes. With the help of a cascaded
LZM2 -IR indicate results with imaginary, real, and combined in- LZM and two overlapping grids, a face recognition scheme that out-
puts, respectively. As expected, preprocessing the input image to performs the Gabor and LBP representations is obtained without any
extract the shape characteristics increases the performance substan- training. We expect that LZM will be successfully used to solve large
tially. The moment orders used in this approach are n1 = 4 for the variety of problems including feature extraction, face/object detec-
first stage, and n2 = 4 for the second stage. Kernel sizes are k1 = 5 tion, and object recognition in addition to face recognition.
and k2 = 7, respectively. The length of the final feature vector is
(N 2 + (N − 1)2 ) × b × K1 × K2 × 2 = 312768 when combining 6. REFERENCES
real and imaginary parts.
[1] Laurenz Wiskott, Jean-Marc Fellous, Norbert Krger, and
Method FaFb FaFc Dup I Dup II Christoph Von Der Malsburg, “Face recognition by elastic
H-LZM 95.0 87.1 73.8 70.9 bunch graph matching,” Pattern Analysis and Machine Intelli-
H-LZM (Weighted) 97.5 95.4 78.5 76.1
gence, vol. 19, pp. 775–779, 1997.
H-LZM2 -I 96.2 97.9 79.6 76.9
H-LZM2 -I (Weighted) 98.7 99.5 83.9 82.5 [2] Baochang Zhang, Shiguang Shan, Xilin Chen, and Wen Gao,
H-LZM2 -R 96.2 96.9 77.4 73.5 “Histogram of Gabor Phase Patterns (HGPP): A Novel Object
H-LZM2 -R (Weighted) 98.7 99.0 83.2 81.2 Representation Approach for Face Recognition,” IEEE Trans-
H-LZM2 -IR 96.3 97.9 79.9 76.5 actions on Image Processing, vol. 16, no. 1, pp. 57–68, 2007.
H-LZM2 -IR (Weighted) 98.7 99.5 84.8 82.5 [3] Timo Ahonen, Abdenour Hadid, and Matti Pietikäinen, “Face
recognition with local binary patterns,” in Proc. 8th Eur. Conf.
Table 2. Recognition rates with different settings. Computer Vision, 2004, pp. 469–481.
[4] Ming-Kuei Hu, “Visual pattern recognition by moment invari-
Finally, we compare our best results with some well-known face ants,” IRE Transactions on Information Theory, vol. 8, no. 2,
recognition approaches in Table 3. Results indicate that the proposed pp. 179–187, 1962.
LZM transformation outperforms the similar methods based on LBP
and Gabor representations. [5] A. Khontanzad and Y. H. Hong, “Rotation invariant image
recognition using features selected via a systematic method,”
Method FaFb FaFc Dup I Dup II Pattern Recogn., vol. 23, pp. 1089–1101, 1990.
LBP [3] 93.0 51.0 61.0 50.0 [6] Atsushi Ono, “Face recognition with zernike moments,” Sys-
LBP (Weighted) [3] 97.0 79.0 66.0 64.0 tems and Computers in Japan, vol. 34, no. 10, pp. 26–35, 2003.
HGPP [2] 97.6 98.9 77.7 76.1
[7] Chandan Singh, Neerja Mittal, and Ekta Walia, “Face recog-
HGPP (Weighted) [2] 97.5 99.5 79.5 77.8
H-LZM2 -IR 96.3 97.9 79.9 76.5
nition using zernike and complex zernike moment features,”
H-LZM2 -IR (Weighted) 98.7 99.5 84.8 82.5 Pattern Recognition and Image Analysis, vol. 21, pp. 71–81,
2011.
Table 3. Recognition rates compared to other methods. [8] Javad Haddadnia, Majid Ahmadi, and Karim Faez, “An effi-
cient feature extraction method with pseudo-zernike moment
in rbf neural network-based human face recognition system,”
EURASIP J. Appl. Signal Process., vol. 2003, pp. 890–901,
5. CONCLUDING REMARKS January 2003.
In this paper, we have introduced a new representation called LZM [9] Hamidreza Rashidy Kanan, Karim Faez, and Yongsheng Gao,
for face recognition by computing Zernike moments at every pixel “Face recognition using adaptively weighted patch pzm array
of an face image using the local neighborhood of each pixel. In this from a single exemplar image per person,” Pattern Recogni-
representation, a face image is decomposed into a set of complex tion, vol. 41, no. 12, pp. 3799 – 3812, 2008.
images corresponding to the moment components, each of which [10] Syed A Rizvi, P Jonathon Phillips, and Hyeonjoon Moon, “The
exposes different characteristics of the face image by capturing the feret verification testing protocol for face recognition algo-
micro structure around each pixel. We have demonstrated that the rithms,” Computer Engineering, , no. October, pp. 48, 1998.
588