ff6 fsc fc0 sc0 ls0">
, jif, i: ini-
tial, f: final.
For the GPC ain represented in the figure below (Fig-
ure 13) we calculate the equations of the first EPC and
the last EPC in the tested GPC, and then we check if
there is intersection between the two EPCs (Pi: intersect-
tion point). Thus we look for the existence of two inflex-
ion points P1 and P2 where there is a sudden change of
the angle deviation of the EPCs forming the GPC. For
example, we have a set of EPCs Right oblique shaft and
a set of EPCs valley after the P1 point and a last set of
EPCs Left oblique shaft after point P2.
5. Refining
After the execution of the ten genetic algorithms (a ge-
netic algorithm for each GPC) we used a process of re-
finement of the results obtained after this execution.
In this stage we favor complex GPCs (Ain, Occlusion
an half occlusions) to simple GPC. In fact, we firstly d
Copyright © 2012 SciRes. JILSA
Genetic Algorithms for Perceptual Codes Extraction
Copyright © 2012 SciRes. JILSA
262
Figure 11. Search of GPC (Right oblique shaft and Left oblique shaft) in the EPCs matrix of the letter “”.
Shaft
Leftoblique shaft
Right oblique
shaft
Figure 12. Extraction of GPC Right half opening occlu-
sion. (a)
Up half opening
occlusion
(b)
Figure 13. Extraction of GPC Ain.
chose the GPCs Ain, Occlusion and subsequent half oc-
clusion GPCs (Left half occlusion, Right half occlusion,
Down half occlusion and Up half occlusion) and we
complete the simple GPCs (Shaft, valley, Left oblique
shaft and Right oblique shaft). Figures 14(a) and (b) re-
present the results of the letter “W” generated by the ex-
tractor of the global perceptual codes. For the first re-
sult, genetic algorithms generate continued following
GPCs: Shaft, Left oblique shaft, Right oblique shaft and
Left oblique shaft. For the second result genetic algori-
thms generate the following GPCs: two Up half opening
occlusion.
(b)
Figure 14. (a) First result for the GPCs forming the letter
“W”; (b) Second result for the GPCs forming the letter
“W”.
To determine the final set of GPCs forming the letter
“W”, we use the stage of refinement already introduced.
Knowing that the final refinement favors complex GPC
(Ain, Occlusion and half occlusions) compared to other
GPCs (Shaft, valley, Left oblique shaft and Right oblique
Genetic Algorithms for Perceptual Codes Extraction 263
shaft), the final result generated by the GPC extractor
(after refinement) is: two Up half opening occlusions.
6. Simulation Results
In the following part we present some examples of EPCs
extraction and then GPCs extraction. Figure 15(a) repre-
sents the generation of the capital letter “D” by EPCs.
The EPC extractor has detected 13 EPCs forming the
letter “D”. Using the EPCs as inputs, the GPC extractor
generated several GPCs building the letter “D”. A re-
finement stage is necessary to maintain the better ones
(GPC 1 = Shaft and GPC 6 = Left half opening occlusion)
(see Figure 15(b)). By the same way, the Arabic letter
”, the French word “un” and the Arabic word “
was generated by the EPCs (see respectively Figures
16(a), 17(a) and 18(a)), and then theses scripts (“”,
“un” and “”) was generated by the GPCs (see
respectively Figures 16(b) , 17(b) and 18(b)).
Referring the simulation examples, we can note the
reduction of the perceptual codes representing a script
using the GPCs comparing to case when the EPCs are
adopted. An other interesting advantage is that the use of
GPCs provides much more significant representation of
the script.
3
2
1
2
3
24
2
4
1
1
1
1
(a)
16
(b)
Figure 15: (a) Generation of the letter “D” with the EPCs;
(b) Generation of the letter “D” with the GPCs.
3
3
2
4
4
1
1
3
1
1
1
3
2
4
4
4
3
1
4
42
3
1
(a)
(a)
7
7
7
(b)
Figure 16. (a) Generation of the arabic letter “” with the
EPCs; (b) Generation of the arabic letter “” with the
GPCs.
3
3
3
1
1
1
1
1
1
1
1
3
3
2
4
2
3
3
3
2
4
1
1
3
3
1
1
1
4
4
2
4
1
1
1
4
(a)
(a)
38
81
7
(b)
Figure 17. (a) Generation of the word “un” with the EPCs;
(b) Generation of the word “un” with the GPCs.
Copyright © 2012 SciRes. JILSA
Genetic Algorithms for Perceptual Codes Extraction
264
1
3
3
4
4
41
1
1
4
4
1
22
22
2
4
4
22
4
42
3
3
1
4
4
3
4
2
4
3
(a)
7
27
10
2
4
7
(b)
Figure 18. (a) Generation of the Arabic word “” with the
EPCs; (b) Generation of the Arabic word “” with the
GPCs.
7. Conclusions
In this paper, we present a new method of features ex-
traction of online handwriting. This method has attem-
pted to overcome the inherent ambiguities of handwriting
with the help of genetic algorithms. It was the difficult
part of the whole handwriting recognition system as the
features extraction had to be robust to cope up with the
handwriting variety and changes due to mood, health and
different writing styles.
To extract the GPCs of an online script we use the
Beta-elliptic model to modelise and to extract parameters
of handwriting. With the help of these parameters we de-
veloped an EPC extractor. For each elliptic arc and with
its deviation angle we define four types of EPCs. The
human visual sense is selectively activated in response to
global form. For this reason we developed a GPC ex-
tractor composed of ten GPCs. A GPC is a combination
of a set of EPCs according to well defined criteria. For
each GPC we used a genetic algorithm to optimize the
choice of a good combination (number and type of EPCs
composing the GPC) of EPCs. Finally a lot of proposi-
tion was giving by the GPC extractor to compose the
script. To choice the best and significant proposition a
stage of refinement was developed.
These GPCs can be used to develop a handwriting
recognition system.
REFERENCES
[1] E. Anquetil and G. Lorette, “Perceptual Model of Hand-
writing Drawing Application to the Handwriting Seg-
mentation Problem,” International Conference on Docu-
ment Analysis and Recognition, Ulm, 18-20 August 1997,
pp. 112-117.
[2] P. Viviani and N. Stucchi, “Biological Movements Look
Uniform: Evidence of Motor-Perceptual Interactions,”
Journal of Experimental Psychology: Human Perception
and Performance, Vol. 18, No. 3, 1992, pp. 603-623.
doi:10.1037/0096-1523.18.3.603
[3] D. Connell Scott, “On Line Handwritten Recognition
Using Multiple Pattern Class Models,” Ph.D. Thesis, Uni-
versité de Michigan state, East Lansing, 2000.
[4] A. L. Koerich, R. Sabourin and C. Y. Suen, “Large Vo-
cabulary Off-Line Handwriting Recognition: A Survey,”
Pattern Analytic Application, Vol. 6, No. 2, 2003, pp. 97-
121.
[5] A. Pavlidis, “36 Years on the Pattern Recognition Front,”
Pattern Recognition Letters, Vol. 24, No. 1-3, 2003, pp.
1-7. doi:10.1016/S0167-8655(02)00211-8
[6] R. Plamondon and S. N. Srihari, “On-Line and Off-Line
Handwriting Recognition: A Comprehensive Survey,”
IEEE Transaction on Pattern Analysis and Machine In-
telligence, Vol. 22, No. 1, 2000, pp. 63-84.
[7] M. Côté, E. Lecolinet, M. Cheriet and C. Y. Suen, “Auto-
matic Reading of Cursive Scripts Using a Reading Model
and Perceptual Concepts, The PERCEPTO System,” In-
ternational Journal on Document Analysis and Recogni-
tion, Vol. 1, No. 1, 1998, pp. 3-17.
doi:10.1007/s100320050002
[8] S. S. Maddouri, “A Neural Perceptual Model Combining
a Local and Global Vision for Arabic Handwritten Words
Recognition,” Ph.D. Thesis, National School of Engi-
neers of Tunis, Tunis, 2002.
[9] M. Côté, “The Use of Access Lexical Model and Percep-
tual Concepts for Image Recognition of Cursive Words,”
Ph.D. Thesis, TELECOM ParisTech, Paris, 1997.
[10] J. R. Pinales, “Recognition of Offline Cursive Handwrit-
ing Using Perceptual and Neural Models,” Ph.D. Thesis,
TELECOM ParisTech, Paris, 2001.
[11] D. Marr, “VISION: A Computational Investigation into
Human Representation and Processing of Visual Infor-
mation,” W. H. Freeman and Company, San Francisco,
1982, pp. 8-41.
[12] J. L. McClelland and D. E. Rumelhart, “Distributed
Memory and the Representation of General and Specific
Information,” Journal of Experimental Psychology: Ge-
neral, Vol. 114, No. 2, 1985, pp. 159-188.
doi:10.1037/0096-3445.114.2.159
[13] Y. Wada, Y. Koike, E. Bateson and M. Kawato, “A
Computational Model for Cursive Handwriting Based on
the Minimization Principle,” Proceeding of the 7th Con-
ference of the Advances in Neural Information Processing
Systems 6, Denver, 29 November-2 December 1993, pp.
727-734.
[14] M. A. Alimi, “Beta Neuro-Fuzzy Systems,” TASK Quar-
terly Journal, Vol. 7, No. 1, 2003, pp. 23-41.
[15] M. A. Alimi, “A Neuro-Fuzzy Approach to Recognize
Arabic Handwritten Characters,” International Confer-
ence on Neural Networks, Houston, 9-12 June 1997, pp.
1397-1400.
[16] H. Bezine, “Contribution to the Development of a Theory
of Handwritten Generation,” Ph.D. Thesis, National School
of Engineers of Sfax, Sfax, 2005.
Copyright © 2012 SciRes. JILSA
Genetic Algorithms for Perceptual Codes Extraction
Copyright © 2012 SciRes. JILSA
265
[17] H. Bezine, M. A. Alimi and N. Sherkat, “Generation and
Analysis of Handwriting Script with the Beta-Elliptic
Model,” Proceedings of the 9th International Workshop
on Frontiers in Handwriting Recognition, Tokyo, 26-29
October 2004, pp. 515-520. doi:10.1109/IWFHR.2004.45
[18] H. Bezine, M. Kefi and M. A. Alimi, “On the Beta-Ellip-
tic Model for the Control of the Human Arm Movement,”
International Journal of Pattern Recognition and Artifi-
cial Intelligence, Vol. 21, No. 1, 2007, pp. 5-19.
doi:10.1142/S0218001407005272
[19] F. L. Chung, T. C. Fu and R. W. P. Luk, “An Evolution-
ary Approach to Pattern-Based Time Series Segmenta-
tion,” IEEE Transactions on Evolutionary Computation,
Vol. 8, No. 5, 2004, pp. 471-489.
doi:10.1109/TEVC.2004.832863
[20] K. Deb, “Genetic Algorithm in Search and Optimization:
The Technique and Applications,” Proceedings of Inter-
national Workshop on Soft Computing and Intelligent
Systems, Calcutta, 12-13 January 1998, pp. 58-87.
[21] S. Forrest, R. E. Smith, B. Jakornik and A. S. et Perelson,
“Using Genetic Algorithms to Explore Pattern Recogni-
tion in the Immune System,” Evolutionary Computation,
Vol. 1, No. 3, 1993, pp. 191-211.
doi:10.1162/evco.1993.1.3.191
[22] C. Gagné and M. Parizeau, “Genetic Engineering of Hi-
erarchical Fuzzy Regional Representations for Handwrit-
ten Character Recognition,” International Journal on
Document Analysis and Recognition, Vol. 8, No. 4, 2006,
pp. 223-231. doi:10.1007/s10032-005-0005-6
[23] J. P. J. Alberto and C. P. C. Juan, “Genetic Algorithms
for Linear Feature Extraction,” Pattern Recognition Let-
ters, Vol. 27, No. 13, 2006, pp. 1508-1514.
doi:10.1016/j.patrec.2006.02.011
[24] M. S. William and A. D. J. Kenneth, “An Analysis of
Multipoint Crossover,” Proceedings of the First Work-
shop on Foundation of Genetic Algorithms, Bloomington,
15-18 July 1990, pp. 301-315.
[25] D. Delahaye, J. M. Alliot, M. Schoenauer and J. L. Farges,
“Genetic Algorithms for Partitioning Air Space,” Pro-
ceedings of the Tenth IEEE Conference on Artificial In-
telligence for Application, San Antonio, 1-4 March 1994,
pp. 291-297.
[26] G. Menier, “On-Line System of Reading and Writing
Cursive Handwriting: Continuous Analysis of Features
and Global Interpretation Optimized by Genetic Algo-
rithm,” Ph.D. Thesis, University Rennes 1, Rennes, 1995.
[27] H. L. Sai and K. L. Hak, “Playing Tic-Tac-Toe Using Ge-
netic Neural Network with Double Transfer Functions,”
Journal of Intelligent Learning Systems and Applications,
Vol. 3, No. 1, 2011, pp. 37-44.
doi:10.4236/jilsa.2011.31005
[28] S. Budi, A. B. Muhammad and E. W. Stefanus, “A Cross
Entropy-Genetic Algorithm for M-Machines No-Wait Job-
Shop Scheduling Problem,” Journal of Intelligent Learn-
ing Systems and Applications, Vol. 3, No. 3, 2011, pp.
171-180. doi:10.4236/jilsa.2011.33018
[29] Z. Bahadır and Ö. brahim, “An Improved Genetic Algo-
rithm for Crew Pairing Optimization,” Journal of Intelli-
gent Learning Systems and Applications, Vol. 4, No. 1,
2012, pp. 70-80. doi:10.4236/jilsa.2012.41007