Beruflich Dokumente
Kultur Dokumente
146
Proceedings of The Fourth International Conference on Informatics & Applications, Takamatsu, Japan, 2015
LBP( xc , yc ) s( g p g c )2 p.
p 0
(1)
147
Proceedings of The Fourth International Conference on Informatics & Applications, Takamatsu, Japan, 2015
=1( )
(2)
148
Proceedings of The Fourth International Conference on Informatics & Applications, Takamatsu, Japan, 2015
RLGS Algorithm:
1. Determine target pixel I(x,y) in the Local
Graph.
2. Construct (LGS) graph G.
3. Compute local standard deviation.
4. Assign 1 If the current pixel >= next
pixel+
5. Move to the next pixel till we finish
processing G.
6. Go to step 3.
7. If we process all the image then compute the
feature histogram h.
The (RLGS), generates a descriptor n; then its
histogram computed h. The resulting histogram
feature encodes the texture pattern.
On the other hand, the algorithm can be further
exploited by defining the distance measure.
Therefore, in this research work we experiment
different distance measures to classify the texture
pattern:
-
(3)
- Chi-square statistics ( 2 ):
2 (, ) =
( )2
(4)
- Correlation coefficient ( ):
(, ) =
( )( )
[( )2 ])([( )2 ])
(5)
=1 ( )
(, ) =
=1
( 2 )
=1
(6)
( 2 )
149
Proceedings of The Fourth International Conference on Informatics & Applications, Takamatsu, Japan, 2015
TABLE 1
BEST CLASSIFICATION RATE OBTAINED BY EACH METHOD
Algorithm
Error
Rate
Distance
Measure
LBP
75.45
24.55
Chi-square
Rotation
invariant
LBP
LPQ
68.18
31.82
Correlation
Coefficient
76.36
23.64
Chai-Square
LGS
75.45
24.55
Chai-Square
SLGS
73.63
26.37
Histogram
Intersection
RLGS
81.81
18.19
Chai-Square
/Histogram
Intersection
V. CONCLUSIONS
In conclusion, this paper extended the idea of the
(LGS) by considering the standard deviation of the
local graph. The aim was to capture more spatial
information. Moreover, chai-square and histogram
intersection distance measures achieved a
significance performance with compare to
correlation coefficient and cosine distance measures.
The obtained results showed a good performance
with compare to other methods such as (LBP, LPQ,
LGS and SLGS). In addition, future work is required
in order to validate the operator against noisy images
and also to look further in applying the method to
other image processing computer vision
applications.
REFERENCES
[1]
CLASSIFICATION RATE
LGS
LPQ
LBP-Uinform
SLGS
RLGS
100
50
0
CHI
COSINE
HIST
CORR
DISTANCE MEASURES
150
Proceedings of The Fourth International Conference on Informatics & Applications, Takamatsu, Japan, 2015
[12] Ojala, T., Pietikinen, M., & Menp, T. (2002).
Multiresolution gray-scale and rotation invariant texture
classification with local binary patterns. Pattern Analysis and
Machine Intelligence, IEEE Transactions on, 24(7), 971-987.
[13] Abusham, E. E., & Bashir, H. K. (2011). Face recognition using
local graph structure (LGS). In Human-Computer Interaction.
Interaction Techniques and Environments (pp. 169-175).
Springer Berlin Heidelberg.
[14] Sayeed, S., Yusof, I., Bashier, H. K., Hossen, J., & Azli, M. A.
(2013). Plant identification based on leaf shape and texture
pattern using local graph structure (LGS). Australian Journal of
Basic and Applied Sciences, 7(11), 29-35.
[15] Bashier H. K., Abusham E. A., and Khalid F., "Face
detection based on graph structure and neural networks." Trends
in Applied Sciences Research 7.8 (2012): 683-691.
[16] Abusham, E. E. A., & Bashier, H. K. (2013). Face recognition
using local graph theory (LGT). In Computing, Electrical and
Electronics Engineering (ICCEEE), 2013 International
Conference on (pp. 593-596). IEEE.
[17] Bashier, H. K., Lau, S. H., Han, P. Y., Ping, L. Y., & Li, C. M.
(2014). Face Spoofing Detection Using Local Graph Structure.
In
2014
International
Conference
on
Computer,
Communications and Information Technology (CCIT 2014).
Atlantis Press.
[18] Tan, X., & Triggs, B. (2010). Enhanced local texture feature sets
for face recognition under difficult lighting conditions. Image
Processing, IEEE Transactions on, 19(6), 1635-1650.
[19] Abdullah, M. F. A., Sayeed, M. S., Muthu, K. S., Bashier, H. K.,
Azman, A., & Ibrahim, S. Z. (2014). Face recognition with
symmetric local graph structure (slgs). Expert Systems with
Applications, 41(14), 6131-6137.
[20] Dey, A., Pradhan, R., Pal, A., & Pal, T. (2015). The Fuzzy
Robust Graph Coloring Problem. In Proceedings of the 3rd
International Conference on Frontiers of Intelligent Computing:
Theory and Applications (FICTA) 2014 (pp. 805-813). Springer
International Publishing.
[21] Shang, Y. (2014). Unveiling robustness and heterogeneity
through
percolation
triggered
by
random-link
breakdown. Physical Review E, 90(3), 032820.
[22] Aguilar, W., Frauel, Y., Escolano, F., Martinez-Perez, M. E.,
Espinosa-Romero, A., & Lozano, M. A. (2009). A robust graph
transformation matching for non-rigid registration. Image and
Vision Computing, 27(7), 897-910.
151