Development of a feature extraction technique for online character recognition system

Innovative Systems Design and Engineering www.iiste.org
ISSN 2222-1727 (Paper) ISSN 2222-2871 (Online)
Vol 3, No 3, 2012

Development of a Feature Extraction Technique for Online
Character Recognition System
Fenwa Olusayo Deborah*
Department of Computer Science and Engineering,
Ladoke Akintola University of Technology, P.M.B 4000, Ogbomoso, Nigeria.
*E-mail of the corresponding author: odfenwa@lautech.edu.ng

Omidiora Elijah Olusayo
E-mail: omidiorasayo@yahoo.co.uk

Fakolujo Olaosebikan Alade
E-mail: ola@fakolujo.com

Abstract
Character recognition has been a popular research area for many years because of its various application
potentials. Some of its application areas are postal automation, bank cheque processing, automatic data
entry, signature verification and so on. Nevertheless, recognition of handwritten characters is a problem
that is currently gathering a lot of attention. It has become a difficult problem because of the high
variability and ambiguity in the character shapes written by individuals. A lot of researchers have proposed
many approaches to solve this complex problem but none has been able to solve the problem completely in
all settings. Some of the problems encountered by researchers include selection of efficient feature
extraction method, long network training time, long recognition time and low recognition accuracy. This
paper developed a feature extraction technique for online character recognition system using hybrid of
geometrical and statistical features. Thus, through the integration of geometrical and statistical features,
insights were gained into new character properties, since these types of features were considered to be
complementary.
Keywords: Character recognition, Feature extraction, Geometrical Feature, Statistical Feature, Character.

1. Introduction
Character recognition is the process of applying pattern-matching methods to character shapes that has
been read into a computer to determine which alpha-numeric character, punctuation marks, and symbols
the shapes represent. The classes of recognition systems that are usually distinguished are online systems
for which handwriting data are captured during the writing process (which makes available the information
on the ordering of the strokes) and offline systems for which recognition takes place on a static image
captured once the writing process is over (Anoop and Anil, 2004; Liu et al., 2004; Mohamad and Zafar,
2004; Naser et al., 2009; Pradeep et al., 2011). The online methods have been shown to be superior to their
offline counterpart in recognising handwriting characters due the temporal information available with the
formal (Pradeep et al., 2011). Handwriting recognition system can further be broken down into two

10

Vol 3, No 3, 2012

categories: writer independent recognition system which recognizes wide range of possible writing styles
and a writer dependent recognition system which recognizes writing styles only from specific users
(Santosh and Nattee, 2009).
Online handwriting recognition today has special interest due to increased usage of the hand held devices.
The incorporation of keyboard being difficult in the hand held devices demands for alternatives, and in this
respect, online method of giving input with stylus is gaining quite popularity (Gupta et al., 2007).
Recognition of handwritten characters with respect to any language is difficult due to variability of writing
styles, state of mood of individuals, multiple patterns to represent a single character, cursive representation
of character and number of disconnected and multi-stroke characters (Shanthi and Duraiswamy, 2007).
Current technology supporting pen-based input devices include: Digital Pen by Logitech, Smart Pad by
Pocket PC, Digital Tablets by Wacom and Tablet PC by Compaq (Manuel and Joaquim, 2001). Although
these systems with handwriting recognition capability are already widely available in the market, further
improvements can be made on the recognition performances for these applications.
The challenges posed by the online handwritten character recognition systems are to increase the
recognition accuracy and to reduce the recognition time (Rejean and Sargurl, 2000; Gupta et. al., 2007).
Various approaches that have been used by many researchers to develop character recognition systems,
these include; template matching approach, statistical approach, structural approach, neural networks
approach and hybrid approach. Hybrid approach (combination of multiple classifiers) has become a very
active area of research recently (Kittler and Roli, 2000; 2001). It has been demonstrated in a number of
applications that using more than a single classifier in a recognition task can lead to a significant
improvement of the system’s overall performance. Hence, hybrid approach seems to be a promising
approach to improve the recognition rate and recognition accuracy of current handwriting recognition
systems (Simon and Horst, 2004). However, Selection of a feature extraction method is probably the single
most important factor in achieving high recognition performance in character recognition system (Pradeep,
Srinivasan and Himavathi, 2011). No matter how sophisticated the classifiers and learning algorithms, poor
feature extraction will always lead to poor system performance (Marc, Alexandre76 and Christian, 2001).

2. Research Methodology
Hundreds of features which are available in the literature can be categorized as follows: Global
transformation and series expansion features, Statistical features and Geometrical and topological features.
Many feature extraction techniques have been proposed in literature to improve overall recognition rate;
however, most of the techniques used only one property of the handwritten character. This research focuses
on developing a feature extraction technique that combined three characteristics (stroke information,
contour pixels and zoning) of the handwritten character to create a global feature vector. Hence, in this
research work, a hybrid feature extraction algorithm was developed to alleviate the problem of poor feature
extraction algorithm of online character recognition system.

2.1 Development of the Proposed Hybrid Feature Extraction Algorithm
The most important aspect of handwriting character recognition scheme is the selection of good feature set
which is reasonably invariant with respect to shape variation caused by various writing styles. Feature
extraction is the process of extracting from the raw data the information which is the most relevant for
classification purposes, in the sense of minimizing the within-class pattern variability while enhancing the
between-class pattern variability (Naser et al., 2009). Features are unique characteristics that can represent
an image, i.e. a character in this case. Each character is represented as a feature vector, which becomes its
identity. The goal of feature extraction is to extract a set of features, which maximizes the recognition rate
with the least amount of elements. Many feature extraction techniques have been proposed to improve
overall recognition rate; however, most of the techniques used only one property of the handwritten
character. This research focuses on a feature extraction technique that combined three characteristics of the
handwritten character to create a global feature vector. A hybrid feature extraction algorithm was
developed using Geometrical and Statistical features as shown in Figure 2.10. Integration of Geometrical

11

Vol 3, No 3, 2012

and Statistical features was used to highlight different character properties, since these types of features are
considered to be complementary (Heute and Paquet, 1998; Cai and Liu, 1998).
2.1.1 Geometrical Features
Various global and local properties of characters can be represented by geometrical and topological
features with high tolerance to distortions and style variations. This type of representation may also encode
some knowledge about the structure of the object or may provide some knowledge as to what sort of
components make up that object. The geometrical features used in this research work were the Stroke
Information and Contour pixel of the characters.

2.1.1.1 Stroke Information
Stroke Information is a combination of local and global features, which are aimed to capture the
geometrical and topological features of the characters and efficiently distinguish and identify the character
from a small subset of characters. Stroke is storage of pen movements in online handwriting recognition.
These movements appear at various positions on view point and joining these positions in first-come-first-
serve basis shows the appearance of drawn text. A character may consist of single or multiple strokes. The
list formed in data collection includes nodes, where each node includes two fields, namely, point and stroke
number. Here, the point represents a coordinate of view point and stroke number represents identity and
sequential order of stroke. Higher recognition performance would be possible if on-line recognition
methods were able to address drawing motion vector (stroke) information (Nishimura and Timikawa,
2003). The feature sets consist of:
(i) Stroke Number
Stroke number helps in identifying similar points, gaps and crossings. The pen movement consists of three
functions, namely, Pen-Down, Pen-Move and Pen-Up. When one presses, moves, lifts the pen up
consecutively, and more than one point collected, the stroke number is incremented. Pen-Move function
stores movements of pen on writing pad. An example of a digital pen for generating stroke information is
as shown in figure 2.1. Figure 2.2 shows a typical example of how different stroke numbers are generated.
However, only stroke is not enough because most of the time different character may get the same no of
strokes. Therefore, in this research, PEN-UP is used as a feature to check how well the character matches
the standard one (i.e. the average for the same character in the database). This feature is calculated by using
the average strokes of a specific character as an input using the membership function as in Equation 2.1:
PEN-UP = e|average - x| (2.1)
where x is the real strokes for the specific character.
(ii) Pressure of the Stroke
This is the pressure representing Pen Ups and Downs in a continuous manner. The use of pen pressure as a
feature is used for the improvement of a basic performance of the writer- independent online character
recognition. The value of the pen pressure exerted on the writing pad was also used as feature. Moreover,
recognition performance could be raised using writing pressure information of on-line writer identification
systems and on-line character recognition systems (Nishimura and Timikawa, 2003).
(iii) Number of Junctions and their Location
A black pixel is considered to be a junction if there are more than two black pixels in its 5 by 7
neighbourhood in the resolution of the character image. The number of junctions as well as their positions
in terms of 35(5x7) quadrants are recorded. For example the character image of Figure 2.3 has 2 junctions
in quadrants 2 and 17. Junctions lying within a pre-defined radial distance are merged into a single junction
and the junctions associated with the headline are ignored.
(iv) Horizontal Projection Count
Horizontal Projection Count is represented as HPC(i) = ∑ , , where F(i,j) is a pixel value (1 for black
background and 0 for white foreground) of a character image, and i and j denote row and column positions

12

Vol 3, No 3, 2012

of a pixel, with the image’s top left corner set to F(0,0). It is calculated by scanning the image row-wise and
finding the sum of background pixels in each row (Figure 2.4). To take care of variations in character sizes,
the horizontal projection count of a character image is represented by percentage instead of an absolute
value and in this present work it is stored as a 4 component vector where the four components symbolize
the percentage of rows with 1 pixel, 2 pixels, 3 pixels and more than 3 pixels. The components of this
vector for the character image given in Figure 2.4 will be [50, 0, 10, 10], as there are 5 rows with 1 pixel;
no rows with 2 pixels; 1 row with 3 pixels and 1 row with more than 3 pixels.

2.1.1.2 Contour Pixels
Correct extraction of the contour will produce more accurate features that will increase the chances of
correctly classifying a given character or pattern. But the question that might arise is why first extract the
contour of a pattern and then collect its features? Why not collect features directly from the pattern? One
answer is, the contour pixels are generally a small subset of the total number of pixels representing a
pattern. Therefore, the amount of computation is greatly reduced when feature extracting algorithms are run
on the contour instead of the whole pattern. Because the contour shares a lot of features with the original
pattern, but has fewer pixels, the feature extraction process becomes much more efficient when performed
on the contour rather on the original pattern. Contour tracing is often a major contributor to the efficiency
of the feature extraction process, which is an essential process in pattern recognition (Liu et al., 2003; Liu
et al., 2004).
In order to extract the contour of the pattern, he following actions must be taken: every time a black pixel is
encountered, turn left, and every time a white pixel is encountered, turn right, until the starting pixel is met
again. All the black pixels traced out is the contour of the pattern. The contour tracing algorithm used in
this research is based on the model developed by Yamaguchi et al. (2003). This is demonstrated in Figure
2.5 (Liu et al., 2003; 2004).
Contour Tracing Algorithm (Yamaguchi et al., 2003):
Input: An image I containing a connected component P of black pixels.
Output: A sequence B (b1, b2, . . . , bk) of boundary pixels, that is, the outer contour.
Begin
Set B to be empty
From bottom to top and left to right scan the cells of I until a black pixel, S, of P is found
Insert S in B
Set the current pixel, P, to be the starting pixel, S
Turn left, that is, visit the left adjacent pixel of P
Update P, that is, set it to be the current pixel
While P not equal to S do
If the current pixel P is black
Insert P in B and turn left (visit the left adjacent pixel of P)
Update P, that is, set it to be the current pixel
Else
Turn right (visit the right adjacent pixel of P)
Update P, that is, set it to be the current pixel.
End

2.1.2 Statistical Features

13

Vol 3, No 3, 2012

Statistical features are derived from the statistical distribution of points. They provide high speed and low
complexity and take care of style variations to some extent. They may also be used for reducing the
dimension of the feature set. The statistical feature adopted in this research is ‘Zoning’. Zone-based feature
extraction method provides good result even when certain pre processing steps like filtering, smoothing and
slant removing are not considered.
Image Centroid and zone-based (ICZ) distance metric feature extraction and Zone Centroid and zone-based
(ZCZ) distance metric feature extraction algorithms were proposed by Vanajah and Rajashekararadhya in
2008 for the recognition of four popular Indian scripts (Kannada, Telugu, Tamil and Malayalam) numerals.
In this research, hybrid of modified Image Centroid and zone-based (ICZ) distance metric feature
extraction and modified Zone Centroid and zone-based (ZCZ) distance metric feature extraction methods
was used. Modifications of the two algorithms are in terms of:
(i) Number of zones being used
(ii) Measurement of the distances from both the Image Centroid and Zone Centroid
(iii) The area of application.

2.1.2.1 The Zoning Algorithm
The most important aspect of handwriting recognition scheme is the selection of good feature set, which is
reasonably invariant with respect to shape variations caused by various writing styles. The zoning method
is used to compute the percentage of black pixel in each zone. The rectangle circumscribing the character is
divided into several overlapping, or non-overlapping regions and the densities of black points within these
regions are computed and used as features as shown in Figure 2.6. The major advantage of this approach
stems from its robustness to small variation, ease of implementation and good recognition rate. Zone-based
feature extraction method provides good result even when certain pre processing steps like filtering,
smoothing and slant removing are not considered. The detailed description of Zoning Algorithm is given as
follows:
The image (character) is further divided in to ‘n’ equal parts (Twenty five in this case) as shown in Figure
2.7. The character centroid (i.e. centre of gravity of the character) is computed and the average distance
from the character centroid to each pixel present in the zone is computed. Similarly zone centroid is
computed and average distance from the zone centroid to each pixel present in the zone is to be computed.
This procedure will be repeated for all the zones/grids/boxes present in the character image. There could be
some zones that are empty, and then the value of that particular zone image value in the feature vector is
zero. Finally, 2 x 25 (i.e. Fifty in this case) such features were used to represent the character image feature.
Algorithm 1 and Algorithm 2 were proposed by Vanajah and Rajashekararadhya in 2008 for the
recognition of four popular Indian scripts (Kannada, Telugu, Tamil and Malayalam) numerals.
Algorithm 1: Image Centroid and Zone-based (ICZ) Distance Metric Feature Extraction System
(Rajashekararadhya and Vanajah, 2008a)
Input: Pre processed numeral image
Output: Features for classification and recognition
Begins
Step 1: Divide the input image in to ‘‘n’’ equal zones
Step 2: Compute the input image centroid
Step 3: Compute the distance between the image centroid to each pixel present in the zone
Step 4: Repeat step 3 for the entire pixel present in the zone
Step 5: Compute average distance between these points
Step 6: Repeat this procedure sequentially for the entire zone
Step 7: Finally, ‘‘n’’ such features will be obtained for classification and recognition

14

Vol 3, No 3, 2012

Ends
Algorithm 2: Zone Centroid and Zone-based (ZCZ) Distance Metric Feature Extraction System
(Rajashekararadhya and Vanajah, 2008b)
Input: Pre processed numeral image
Begins
Step 1: Divide the input image in to n equal zones
Step 2: Compute the zone centroid
Step 3: Compute the distance between the zone centroid to each pixel present in the zone
Step 6: Repeat this procedure sequentially for the entire zone
Step 7: Finally, n such features will be obtained for classification and recognition
Ends

2.1.2.2 Hybrid of the Modified Zoning Feature Extraction Algorithms
The following are the algorithms to show the working procedure of the modified hybrid zoning feature
extraction methods:
Modified Algorithm 1: Image Centroid and Zone-based (ICZ) distance metric feature extraction algorithm.
Input: Pre processed character images
Method Begins
Step 1: Divide the input image in to 25 equal zones as shown in Figure 2.7
Step 2: Compute the input image centroid as shown in Figure 2.8−1
n
using the formula:

m∑ x , where n = width
Centre of gravity in the horizontal direction (x–axis) = −1
n=0
i (2.2)
Centre of gravity in the vertical direction (y–axis) = ∑ y , where m = height
i (2.3)
m =0
Step 3: Compute the distance between the image centroid to each pixel present in the zone as shown in
Figure 2.8
Step 4: Repeat step 3 for the entire pixel present in the zone (five points in this case):
d = d1 + d2 + d3 + d4 + d5 (2.4)
Step 5: Compute average distance between these points as:
Average Image Centroid Distance DI = d/5 (2.5)
0
where d = total distance between the image centroid to the pixel measured at an of angle 20
Step 6: Repeat this procedure sequentially for the entire zone (25 zones).
Total Distance (P) = DI1 + DI2 + DI3 + . . .+ DIm (2.6)
m −1
Total Average Distance j= ∑ DZ
z =0 m (2.7)

where m = 25(total number of zones)
Step 7: Finally, 25 such features was obtained for classification and recognition.
Method Ends.

15

Vol 3, No 3, 2012

Modified Algorithm 2: Zone Centroid and Zone-based (ZCZ) distance metric feature extraction algorithm
Input: Pre processed character image
Method Begins
Step 1: Divide the input image in to 25 equal zones as shown in Figure 2.7
Step 2: Compute the zone centroid for the entire pixel present in the zone as shown in Figure 2.9 using the
formula: n −1

∑ x , where n = width
Centre of gravity in the horizontal direction (x–axis) m −1
=
n =0
i (2.8)
Centre of gravity in the vertical direction (y–axis) = ∑ y , where m = height
i (2.9)
m =0
Step 3: Compute the distance between the zone centroid to each pixel present in the zone
Step 4: Repeat step 3 for the pixel present in a zone (5 points in this case)
Total Distance D = D1 + D2 + D3 + D4 + D5 (2.10)
Step 5: Compute average distance between these points as:
Average distance /5 (2.11)
0
where D = distance between the zone centroid measured at angle 20 to the pixel in the zone
Step 6: Repeat this procedure sequentially for the entire 25 zones.
Total Distance (Q) = DZ1 + DZ2 + DZ3 + . . + DZm (2.12)
m −1
DZ
Total Average Distance k = ∑
z =0 m (2.13)
where m = 25(total number of zones)
Step 7: Finally, 25 such features were obtained for classification and recognition
Method Ends

The Hybrid Zoning Algorithm: Hybrid of Modified ICZ and Modified ZCZ
Input: Pre processed character image
Output: Features for Classification and Recognition
Method Begins
Step 1: Divide the input image into 25 equal zones
Step 2: Compute the input image centroid
Step 3: Compute the distance between the image centroid to each pixel present in the zone
Step 6: Compute the zone centroid
Step 7: Compute the distance between the zone centroid to each pixel present in the zone.
Step 10: Repeat the steps 3-9 sequentially for the entire zones
Step 11: Finally, 2xn (50) such features were obtained for classification and recognition.

16

Vol 3, No 3, 2012

Method Ends

2.2 The Developed Hybrid (Geom-Statistical) Feature Extraction Algorithm
Step 1: Get the stroke information of the input characters from the digitizer (G-pen 450)
These include:
(i) Pressure used in writing the strokes of the characters
(ii) Number (s) of strokes used in writing the characters
(iii) Number of junctions and the location in the written characters
(iv) The horizontal projection count of the character
Step 2: Apply Contour tracing algorithm to trace out the contour of the characters
Step 3: Run Hybrid Zoning algorithm on the contours of the characters
Step 4: Feed the outputs of the extracted features of the characters into the digitization stage in order to
convert all the extracted features into digital forms

3. Conclusion and Future Work
In this medium, we have been able to develop an effective feature extraction technique for online character
recognition system using hybrid of geometrical and statistical features. Precisely, a hybrid feature
extraction was developed to alleviate the problem of poor feature extraction algorithm of online character
recognition system. However, future and further research may be geared towards developing a hybrid of
modified counter propagation and modified optical backpropagation neural network model for the
aforementioned system. Also, the performance of the online character recognition system under
consideration could be evaluated based on learning rates, image sizes and database sizes.

References
Anoop, M. N. and Anil K. J. (2004). Online Handwritten Script Recognition. IEEE Trans. PAMI, 26, 1,
124-130.
Cai, J. and Liu, Z. (1998). Integration of Structural and Statistical information for Unconstrained
Handwritten Numerals Recognition. Proceedings of the 14th International Conference on Pattern
Recognition, 1, 378-380.
Gupta, K., Rao, S.V. and Viswanath (2007). Speeding up Online Character Recognition. Proceedings of
Image and Vision Computing, Newzealand, Hamilton, 41-45.
Heute, L. T. and Paquet, J. V. (1998). A Structural/Statistical feature base Vector for Handwritten
Character Recognition. Pattern Recognition Letters, 19, 629-641.
Kittler, J. and Roli, F. (2000): 1st International Workshop on Multiple Classifier Systems, Cagliari, Italy.
Kittler, J. and Roli, F. (2001): 2nd International Workshop on Multiple Classifier Systems, Cagliari, Italy.
Liu, C. L., Nakashima, K., Sako, H. and Fujisawa, H. (2004). Handwritten Digit Recognition: Investigation
of Normalization and Feature Extraction Techniques. Pattern Recognition, 37, 2, 265-279.
Liu, C. L., Sako, H. and Fujisawa, H (2003). Handwritten Chinese character recognition: Alternatives to
nonlinear normalization, In Proceedings of the 7th International Conference on Document Analysis and
Recognition, Edinburgh, Scotland, 524-528.
Liu, C., Stefan, J. and Masaki N. (2004). Online Recognition of Chinese Characters: The State-of-the- Art.
IEEE Trans. on Pattern Analysis and Machine Intelligence, 26, 2, 198-203.
Manuel, J., Fonseca, and Joaquim, A. J. (2001). Experimental Evolution of an Online Scribble Recognizer.
Pattern Recognition Letters, 22,12,1311-1319.

17

Vol 3, No 3, 2012

Mohamad, D. and Zafar, M. F. (2004). Comparative Study of Two Novel Feature Vectors for Complex
Image Matching Using Counter propagation Neural Network. Journal of Information Technology, FSKSM,
UTM, 16, 1, 2073-2081.
Naser, M. A., Adnan, M., Arefin, T. M., Golam, S. M. and Naushad, A. (2009). Comparative Analysis of
Radon and Fan-beam based Feature Extraction Techniques for Bangla Character Recognition, IJCSNS
International Journal of Computer Science and Network Security, 9, 9, 120-135.
Nishimura, H. and Timikawa, T. (2003). Offline Character Recognition using Online Character Writing
Information. Proceedings of the seventh IEEE International Conference on Document Analysis and
Recognition.
Pradeep, J., Srinivasan, E. and Himavathi, S. (2011). Diagonal Based Feature Extraction for Handwritten
Alphabets Recognition using Neural Network. International Journal of Computer Science and Information
Technology (IJCS11), 3, 1, 27-37.
Rajashekararadhya, S. V. and Vanaja, P. R. (2008a). Handwritten numeral recognition of three popular
South Indian Scripts: A Novel Approach. Proceedings of the Second International Conference on
information processing ICIP, 162-167.
Rajashekararadhya S. V. and Vanaja, P. R (2008b). Isolated Handwritten Kannada Digit Recognition: A
Novel Approach. Proceedings of the International Conference on Cognition and Recognition, 134-140.
Rejean, P. and Sargur, S. N. (2000). On-line and Off-line Recognition: A Comprehensive Survey. IEEE
Transaction on Pattern Analysis and Machine Intelligence, 22, 1, 63-84.
Santosh, K.C. and Nattee, C. (2009). A Comprehensive Survey on Online Handwriting Recognition
Technology and Its Real Application to the Nepalese Natural Handwriting. Kathmandu University Journal
of Science, Engineering Technology, 5, 1, 31-55.
Shanthi, N., and Duraiwamy, K. (2007). Performance Comparison of Different Image Sizes for
Recognizing Unconstrained Handwritten Tamil Characters Using SVM. Journal of Science, 3, 9, 760-764.
Simon G. and Horst B. (2004). Feature Selection Algorithms for the Generalization of Multiple Classifier
Systems and their Application to Handwritten Word Recognition. Pattern Recognition Letters, 25,
11,1323-1336.
Yamaguchi, T., Nakano, Y., Maruyama, M., Miyao H. and Hananoi T. (2003). Digit Classification on
Signboards for Telephone Number Recognition. In Proceedings of the 7th International Conference on
Document Analysis and Recognition, Edinburgh, Scotland, 359–363.

Figure 2.1: The snapshot of Genius Pen (G-Pen 450) Digitizer for Character Acquisition

18

Vol 3, No 3, 2012

Figure 2.2: Writing character “A’’ with 3 Strokes

Figure 2.3: Division of character image into 35 quadrants

19

Vol 3, No 3, 2012

Figure 2.4: Horizontal Projection Count of character image ‘‘F’’

Figure 2.5: The Contour-tracing Algorithm (Yamaguchi et al., 2003)

20

Vol 3, No 3, 2012

Figure 2.6: Feature Extraction using Zoning

Figure 2.7: Character ‘‘n’’ in 5 by 5 (25 equal zones)

21

Vol 3, No 3, 2012

Figure 2.8: Image Centroid of character ‘‘n’’ in zoning

Figure 2.9: Zone Centroid of character ‘‘n’’ in zoning

22

Vol 3, No 3, 2012

Figure 2.10: The Developed Hybrid Feature Extraction Model

23

Development of a feature extraction technique for online character recognition system

More Related Content

What's hot (17)

Similar to Development of a feature extraction technique for online character recognition system (20)

More from Alexander Decker (20)

Recently uploaded (20)

Development of a feature extraction technique for online character recognition system