A Hybrid Color and Frequency Features Method for Face Recognition

paraderollΤεχνίτη Νοημοσύνη και Ρομποτική

17 Νοε 2013 (πριν από 3 χρόνια και 6 μήνες)

54 εμφανίσεις

Zhiming

Liu and
Chengjun

Liu
,
IEEE


Introduction



Algorithms and
procedures



Experiments



Conclusion


Uncontrolled
illumination



Color
information and frequency features derived
by means of the Discrete Fourier Transform (DFT)
help improve face recognition
performance.



This
correspondence presents a novel hybrid Color
and Frequency Features (CFF) method for face
recognition.



RIQ
: R of the
RGB


+

IQ

of the
YIQ

Initial
image

R, I, Q

Real part

Discrete Fourier transform of a face image

Imaginary
part

Magnitude


Mask



EFM method
:


PCA reduce dimensionality


FLD


separated pattern classes


PCA
:



Let
X

𝑅
𝑁




Y
=

[


1

2


𝑚
]
𝑡
𝑋

……..(1)




1

2


𝑚
(m<N)




X :a random vector representing a frequency pattern vector.

: are the eigenvectors



largest eigenvalues of X


FLD:


S
(w)

: within
-
class scatter matrix

S
(b)

: between
-
class scatter matrix,

)
(
)
(
)
(
w
b
S
S
tr
J


Database:
FRGC version 2 Experiment
4



the Training set contains 12 776 images that are either
controlled or uncontrolled.

0.9

0.55

0.95


Future research will consider applying kernel methods,
such as the multiclass Kernel Fisher Analysis (KFA)
method presented in
[9],
to replace the EFM method
for improving face recognition performance
.



The
hybrid color space improves face recognition
performance
.

Thanks for your
attention !!!