DOI : 10.17577/IJERTV2IS70344
- Open Access
- Total Downloads : 261
- Authors : Deepika Rajurkar, Prof. Alwin Anuse
- Paper ID : IJERTV2IS70344
- Volume & Issue : Volume 02, Issue 07 (July 2013)
- Published (First Online): 12-07-2013
- ISSN (Online) : 2278-0181
- Publisher Name : IJERT
- License:
This work is licensed under a Creative Commons Attribution 4.0 International License
Transform Domain Methods For Devangri Character Recognition
Deepika Rajurkar Prof. AlwinAnuse
Abstract
Rapidly growing computational power enables the implementation of Devnagri Character Recognition methodologies. Digital document processing is gaining popularity for application to office and library automation, bank and postal services, publishing houses and communication technology. Character recognition system consists of three steps:
(i) pre-processing (ii) feature extraction and (ii) classification. Features extracted, from the characters using transform domain methods are, Fourier Transform and Wavelet Transform. Three types of classifiers i.e Feed- forward Backpropagation, Generalized Regression Neural Network and K-Nearest Neighbor method are implemented in this paper. Results in terms of recognition rate using combinations of feature extraction methods and classifiers are compared and analysed.
-
Introduction
Artificial Intelligence, giving machines the human like abilities, has remained one of the most challenging areas in Electronic sciences in last few decades. Giving machine the power to see, interpret and the ability to read text is one of the major tasks of AI. A good text recognizer has many commercial and practical applications, e.g. from searching data in scanned book to automation of any organization, like post office, which involve manual task of interpreting text. Devnagri script consists of 11 vowels and 33 consonants. In recognition of Devnagri characters recognition of different handwritten characters are done. Features are extracted from different handwritten and then recognition is done on the basis of extracted features. In Devnagri script some of the characters are similar and therefore recognition of these characters is difficult.Extensive research works on recognition of handwritten characters, have been carried out since the last few decades because it is a classical problem of pattern recognition and it has enormous application potentials. A wide variety of
algorithms / schemes exist for recognition of isolated handwritten characters and each of these has their own merits and demerits. The existing character recognition schemes differ from each other mainly with respect to pre-processing steps used, feature selection / extraction method and classifier used. Combining various pre-processing steps, feature extraction methods, and different classifiers high recognition rate can be achieved. Here, results, in terms of recognition rate, of different pre processing steps and feature extraction methods are compared and discussed [1].
-
Pre-processing
2.12-D Median-filtering
Median filtering is a nonlinear operation often used in image processing to reduce "salt and pepper" noise. A median filter is more effective than convolution when the goal is to simultaneously reduce noise and preserve edges. It performs median filtering of the matrix A in two dimensions. Each output pixel contains the median value in the m-by-n neighborhood around the corresponding pixel in the input image [2].
-
Binarization
Global image threshold using Otsu's method. It computes a global threshold (level) that can be used to convert an intensity image to a binary image [6].
-
Morphological operations
Morphology is a broad set of image processing operations that process images based on shapes. Morphological operations apply a structuring element to an input image, creating an output image of the same size. In a morphological operation, the value of each pixel in the output image is based on a comparison of the corresponding pixel in the input image with its neighbors. By choosing the size and shape of the neighborhood, you can construct a morphological operation that is sensitive to specific shapes in the input image. The most basic morphological operations are dilation and erosion. Dilation adds pixels to the boundaries of objects in an
image, while erosion removes pixels on object boundaries.[6].
-
Cropping
All the characters are uniformly sized and cropped so that further operations can be done only on concern character pixels leaving extra pixels aside.
-
Thinning
Thinning thins objects to lines. It removes pixels so that an object without holes shrinks to a minimally connected stroke, and an object with holes shrinks to a connected ring halfway between each hole and the outer boundary.
-
Segmentation
In most of the existing systems recognition accuracy is heavily dependent on the quality of the input document. In handwritten text adjacent characters tend to be touched or overlapped. Therefore it is essential to segment a given string correctly into its character component[2].
-
-
Feature Extraction
-
Fourier Descriptors Algorithm
The purpose of the algorithm is to extract the information of the boundary (contour) of a segmented character and present it in a more compact form. There are various methods of boundary tracing. The four-neighbor adjacent method is adopted in this system [3].
Figure. 1. Four-neighbor searching sequence
Algorithm:
-
Remove noise by applying filter.
-
Convert gray scale image into binary image.
-
Remove noise (eg. Spikes) by applying erode and dilate morphological operations respectively.
-
To bring uniformity among the numerals, fit a bounding box to the character image, crop, and resize it to a size of 100x 100 pixels.
-
Apply thinning to the character image.
-
Segment the character image into 25 blocks of size 4 x 4 size.
-
Extract the boundary of the character image and resample the boundary in order to obtain a uniform resampling along the running arc length of the boundary.
-
Represent the boundary in the complex plane where the column-coordinate is the real part and the row-coordinate the imaginary part.
-
Compute Fourier transform and obtain the invariant 32 dimensional Fourier descriptors.
-
Calculate magnitude and phase of these 32 fourier descriptors.
-
Nullify the dc value of fourier descriptors as it contains very less information.
-
Calculate standard deviation of 31 phase values.
-
Combining 31 magnitude values and 1 standard deviation value of 31 phase values, total features are 32.
-
Normalize these 32 feature values and apply these to different classifiers to classify the characters.
-
-
Discrete Wavelets Algorithm
-
Remove noise by applying filter.
-
Convert gray scale image into binary image.
-
Remove noise (eg. Spikes) by applying erode and dilate morphological operations respectively.
-
To bring uniformity among the numerals, fit a bounding box to the character image, crop, and resize it to a size of 100x 100 pixels.
-
Apply thinning to the character image.
-
Segment the character image into 25 blocks of size 4 x 4 size.
-
Apply daubechies wavelet transform on the segmented image and calculate discrete wavelets for the segmented image.
-
Normalize these feature values and apply these to different classifiers to classify the characters.
-
-
Fusion (Fourier Descriptors + Discerete Wavelets)
-
In case of fusion, combination of both the features i.e fourier descriptors and discrete wavelets, are used as extracted features.
-
These features arethen normalized before applying to different classifiers to classify the characters.
-
-
-
Classification
-
Feed-forward Backpropagation Algorithm
-
Initialize all Wij s to small random values with Wij being the value of the connection weight between unit j and unit i in the layer below.
-
Present the input vectors i.e. magnitude and phase. Specify the desired outputs. If the net is used as a classifier then all desired outputs are typically set to zero except for that corresponding to the class the input is from.
-
Calculate the outputs yj of all the nodes using the present value of Wij where Wij is the value of connection weight between unit j and the unit i in the layer below:
yj =[1/{1+exp(- yi Wij)}] ….Eqn(1)
This particular nonlinear function is called a sigmoid function.
-
Adjust weights by:
Wij(n+1)= Wij(n)+j yi+(Wij(n)Wij(n-1))
.Eqn (2)
where 0<<1
where (n+l), (n) and (n-1) index next, present and previous, respectively. The parameter is a learning rate similar to step size in gradient search algorithms, and is a constant between 0 and 1 which determines the effect of past weight changes on the current direction of movement in weight space. j, is an error term for node j. If node j is an output node, dj and yi stand for respectively, the desired and actual value of a node, then
j=(dj- yj ) yj(1- yj ) ….Eqn (3)
If node j is an internal hidden node, then :
j= yj(1- yj kWjk ) ….Eqn (4)
where k is over all nodes in the layer above node j.
-
Present another input and go back to step (2). All the training inputs are presented cyclically until weights stabilize (converge).
Following are specifications of the classifier used for training:
-
Learning rate=0.05
-
Epochs=50000
-
Training parameter=50
-
Goal=1e-1
-
3 layers: 1input layer, 1 hidden layer , 1 output layer.
-
Hidden layer uses function tangent sigmoid and output layer pure linear.
-
Levenberg-Marquardt backpropagation recognition engine is used.
-
-
-
Generalized Regression Neural Network Algorithm
-
Calculate the statistically most likely value of each output from a given set of inputs, all taken from the training set. This most likely value is the conditional mean.
-
The algorithm requires the joint probability density function of the inputs and outputs, each treated as random variables. The calculation is done for each output individually. The joint probability density function of vector x inputs and y output may be written as
.Eqn(5)
.Eqn(6)
where, N is the number of input nodes, Ntrain is total number of training sets, is the width parameter subject to certain constraints, and dj is the Euclidean distance between an input and the mean of those inputs.
-
The statistically most likely value for the random variable y (output) given (the vector of random input variables) is
estimated substituting Eqn(5) and Eqn(6) into
.Eqn(7)
j
j
-
The algorithm uses each of the training vectors as the centre of a Gaussian function defined as exp(-D 2/22) . The larger the number of training samples, the larger is the amount of memory required and the longer is the time for the calculation.
-
-
K-Nearest Neighbor Algorithm
-
Form the training dataset P.
-
Given a query instance to be classified x.
-
Let x1, x2,.,xk denote the k instances from training dataset that are nearest to x
-
Return the class that represents the maximum of the k instances.
-
Distance can be calculated from different methods i.e Euclidean, Cityblock, Hamming etc.
-
By default it calculates distance using Euclidean method.
-
Values of k varies in odd sequence i.e k=1,3,5.. Default value is 1.
-
-
Testing and results
-
Database
Database is taken from Indian Statistical Institute, Kolkata. Database consists of 33 consonants. 10 and 20 samples are taken of each character to form the sets of database. In this way three types of database sets are formed:
-
Training sample = 660; Testing sample = 330
-
Training sample = 660 ; Testing sample =330
-
Training sample = 330 ; Testing sample = 660
-
-
Results of Pre-processing
Original images
Figure 2. Original Images
Median filtered images
Figure 3. Median Filtered Images
Dilated and Eroded images
Figure 4. Dilated and Eroded Images
Cropped images
Figure 5. Cropped Images
Thinned images
Figure 6. Thinned Images
Segmented images
Figure 7. Segmented Images
-
Comparison of recognition rate
Table 1- Comparison of recognition rate using database (a)
FEATURE
EXTRACTION
RECOGNITION RATE
METHODS
Fourier
Discrete
Fusion
Descriptors
Wavelets
(Fourier
Descriptor
TYPES OF
+ Discrete
CLASSIFIERS
Wavelets)
Feed-Forward Backpropagation Network
23%
25.4545%
20.9091%
Generalized Regression Neural Network
71.5152%
72.1212%
68.4848%
K-Nearest Neighbour
71.5152%
81.8182%
74.2424%
FEATURE
EXTRACTION
RECOGNITION RATE
METHODS
Fourier
Discrete
Fusion
Descriptors
Wavelets
(Fourier
Descriptor
TYPES OF
+ Discrete
CLASSIFIERS
Wavelets)
Feed-Forward Backpropagation Network
30%
31.0606%
29.0909%
Generalized Regression Neural Network
56.3636%
57.8788%
56.8182%
K-Nearest Neighbour
56.6667%
71.9697%
63.1818%
FEATURE
EXTRACTION
RECOGNITION RATE
METHODS
Fourier
Discrete
Fusion
Descriptors
Wavelets
(Fourier
Descriptor
TYPES OF
+ Discrete
CLASSIFIERS
Wavelets)
Feed-Forward Backpropagation Network
23%
25.4545%
20.9091%
Generalized Regression Neural Network
71.5152%
72.1212%
68.4848%
K-Nearest Neighbour
71.5152%
81.8182%
74.2424%
FEATURE
EXTRACTION
RECOGNITION RATE
METHODS
Fourier
Discrete
Fusion
Descriptors
Wavelets
(Fourier
Descriptor
TYPES OF
+ Discrete
CLASSIFIERS
Wavelets)
Feed-Forward Backpropagation Network
30%
31.0606%
29.0909%
Generalized Regression Neural Network
56.3636%
57.8788%
56.8182%
K-Nearest Neighbour
56.6667%
71.9697%
63.1818%
Graph 1-Graphical Representation of Comparison of Recognition Rates usingDatabase (a)
Table 2- Comparison of recognition rate using database (b)
Graph 2-Graphical Representation of Comparison of Recognition Rates usingDatabase (b)
Table 3- Comparison of recognition rate using database (c)
FEATURE
EXTRACTION
RECOGNITION RATE
METHODS
Fourier
Discrete
Fusion
Descriptors
Wavelets
(Fourier
Descriptor
TYPES OF
+ Discrete
CLASSIFIERS
Wavelets)
Feed-Forward Backpropagation Network
23.9191%
26.8182%
24.6697%
Generalized Regression Neural Network
43.7879%
43.6364%
45.6061%
K-Nearest Neighbour
44.3939%
61.2121%
51.6667%
Graph 3-Graphical Representation of Comparison of Recognition Rates usingDatabase
(c).
-
Confusion Matrix
In the field of machine learning, a confusion matrix is a specific table layout that allows visualization of the performance of an algorithm. Each column of the matrix represents the instances in a predicted class, while each row represents the
instances in an actual class. The name stems from the fact that it makes it easy to see if the system is confusing two classes i.e. commonly mislabeling one as another.Here, confusion matrices are shown of system consists of Discrete Wavelets and KNN of all the 3 database.
Table 4-Confusion Matrix of Discrete Wavelets using KNN with Database (a)
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
1
0
0
0
0
0
0
8
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
7
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
7
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
1
0
0
1
0
0
0
0
0
0
0
0
0
8
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
8
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
7
0
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
9
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0 0
0
0
0
0
1
0
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
7
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
9
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
9
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
8
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
9
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
7
0
0
0
2
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
9
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
8
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
9
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
6
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
7
0
0
1
0
0
1
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
7
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
7
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
9
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
10
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
9
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
10
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
9
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
9
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
10
Table 5-Confusion Matrix of Discrete Wavelets using KNN with Database (b)
11
0
1
1
0
0
1
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
1
0
1
1
0
1
0
0
0
0
14
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
2
0
1
0
0
0
0
11
1
0
0
0
0
1
0
0
0
0
1
0
0
0
0
0
4
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
11
1
0
0
0
0
1
0
0
0
0
1
2
0
0
0
2
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
1
1
11
0
0
0
0
0
0
0
0
0
0
0
1
0
0
2
0
0
0
0
3
0
0
1
0
0
0
0
0
0
0
0
0
1
13
0
1
0
0
0
1
0
0
2
0
0
0
1
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
17
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
16
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
0
1
1
0
0
0
0
0
0
0
1
0
0
0
14
0
0
0
0
0
0
0
0
0
0
3
0
0
0
0
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
18
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
17
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
2
14
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
2
1
1
13
0
1
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
2
1
0
0
0
0
0
0
0
0
0
15
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
13
0
0
0
0
1
0
0
0
0
0
0
3
1
0
0
0
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
13
0
0
0
1
0
0
0
0
3
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
15
0
0
1
0
0
0
0
1
1
0
2
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
16
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
0
0
1
0
0
0
11
0
0
1
0
0
0
0
5
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
14
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
18
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
16
0
0
1
0
0
1
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
13
1
0
0
0
0
0
0
2
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
2
0
0
0
15
1
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
16
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
2
17
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
18
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
2
0
0
1
0
1
0
0
1
1
0
12
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
18
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4
0
0
0
0
1
0
0
0
0
14
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
2
1
1
0
0
1
0
13
0
0
0
0
0
0
0
0
0
0
0
1
1
2
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
14
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
1
2
1
0
0
0
0
0
0
14
9
0
1
1
0
0
1
0
0
0
1
0
0
0
1
1
1
0
0
0
0
0
0
0
0
1
1
1
0
1
0
0
0
0
12
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
4
1
0
0
0
0
1
0
0
0
0
7
1
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
5
0
0
0
2
2
0
0
0
0
0
1
0
0
0
0
0
7
0
1
0
0
0
1
0
0
0
0
1
2
1
0
0
0
0
0
1
1
5
0
0
0
0
0
0
0
0
0
0
1
1
11
0
0
0
0
0
0
0
0
0
1
0
1
0
0
1
0
0
0
0
3
0
0
1
0
0
0
0
0
0
0
0
0
1
9
0
1
0
0
0
1
0
0
2
0
0
1
1
1
0
0
0
0
1
1
0
1
0
0
0
0
0
0
0
0
0
0
0
13
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
4
1
0
0
0
0
0
0
0
0
0
0
0
0
13
/td>
0
0
0
0
0
0
0
1
0
2
0
1
0
0
0
0
0
1
0
0
1
1
0
0
0
0
0
0
1
1
0
0
0
10
0
0
0
0
0
0
0
0
0
0
6
0
0
0
0
0
0
1
0
0
1
0
0
0
0
0
1
0
0
0
0
0
0
16
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
5
13
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
16
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
1
2
13
0
1
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
2
1
0
0
0
0
0
0
0
0
0
15
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
1
0
0
0
2
0
0
0
1
0
0
0
11
0
0
0
0
1
0
0
0
0
0
0
4
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
1
0
8
0
0
0
0
0
0
0
0
6
1
0
0
1
0
1
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
13
0
0
0
0
0
0
0
1
2
0
3
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
1
0
16
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
1
0
0
0
10
0
0
1
0
0
0
0
6
0
0
0
0
0
0
0
0
2
2
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
11
0
0
0
0
0
3
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
6
12
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
15
0
0
1
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
1
0
1
0
0
11
3
0
0
0
0
0
0
2
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
12
1
1
0
1
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
16
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
1
0
0
0
2
15
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
17
0
0
0
0
0
0
0
0
0
2
1
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
1
0
0
0
0
0
14
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
1
0
0
15
0
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
0
0
0
0
1
0
1
1
0
13
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
2
1
1
0
0
1
0
11
0
0
0
0
0
0
0
0
1
0
0
1
1
5
1
0
1
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
8
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
1
1
0
0
1
2
1
0
0
0
0
0
0
12
11
0
1
1
0
0/p>
1
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
1
0
1
1
0
1
0
0
0
0
14
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
2
0
1
0
0
0
0
11
1
0
0
0
0
1
0
0
0
0
1
0
0
0
0
0
4
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
11
1
0
0
0
0
1
0
0
0
0
1
2
0
0
0
2
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
1
1
11
0
0
0
0
0
0
0
0
0
0
0
1
0
0
2
0
0
0
0
3
0
0
1
0
0
0
0
0
0
0
0
0
1
13
0
1
0
0
0
1
0
0
2
0
0
0
1
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
17
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
16
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
0
1
1
0
0
0
0
0
0
0
1
0
0
0
14
0
0
0
0
0
0
0
0
0
0
3
0
0
0
0
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
18
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
17
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
2
14
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
2
1
1
13
0
1
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
2
1
0
0
0
0
0
0
0
0
0
15
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
13
0
0
0
0
1
0
0
0
0
0
0
3
1
0
0
0
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
13
0
0
0
1
0
0
0
0
3
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
15
0
0
1
0
0
0
0
1
1
0
2
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
0
0
16
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
0
0
0
1
0
0
0
11
0
0
1
0
0
0
0
5
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
14
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
18
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
16
0
0
1
0
0
1
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
13
1
0
0
0
0
0
0
2
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
2
0
0
0
15
1
0
0
0
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
16
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
2
17
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
18
0
0
0
0
0
0
0
0
0
0
p>1 0
0
0
0
0
0
0
0
0
1
0
2
0
0
1
0
1
0
0
1
1
0
12
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
18
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4
0
0
0
0
1
0
0
0
0
14
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
2
1
1
0
0
1
0
13
0
0
0
0
0
0
0
0
0
0
0
1
1
2
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
14
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
1
2
1
0
0
0
0
0
0
14
9
0
1
1
0
0
1
0
0
0
1
0
0
0
1
1
1
0
0
0
0
0
0
0
0
1
1
1
0
1
0
0
0
0
12
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
4
1
0
0
0
0
1
0
0
0
0
7
1
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
5
0
0
0
2
2
0
0
0
0
0
1
0
0
0
0
0
7
0
1
0
0
0
1
0
0
0
0
1
2
1
0
0
0
0
0
1
1
5
0
0
0
0
0
0
0
0
0
0
1
1
11
0
0
0
0
0
0
0
0
0
1
0
1
0
0
1
0
0
0
0
3
0
0
1
0
0
0
0
0
0
0
0
0
1
9
0
1
0
0
0
1
0
0
2
0
0
1
1
1
0
0
0
0
1
1
0
1
0
0
0
0
0
0
0
0
0
0
0
13
0
1
0
0
0
0
0
1
0
0
0
0
0
td>
0
0
0
0
0
0
4
1
0
0
0
0
0
0
0
0
0
0
0
0
13
0
0
0
0
0
0
0
1
0
2
0
1
0
0
0
0
0
1
0
0
1
1
0
0
0
0
0
0
1
1
0
0
0
10
0
0
0
0
0
0
0
0
0
0
6
0
0
0
0
0
0
1
0
0
1
0
0
0
0
0
1
0
0
0
0
0
0
16
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
5
13
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
16
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
1
2
13
0
1
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
2
1
0
0
0
0
0
0
0
0
0
15
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
1
0
0
0
2
0
0
0
1
0
0
0
11
0
0
0
0
1
0
0
0
0
0
0
4
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
1
0
8
0
0
0
0
0
0
0
0
6
1
0
0
1
0
1
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
13
0
0
0
0
0
0
0
1
2
0
3
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
1
0
16
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
1
0
0
0
10
0
0
1
0
0
0
0
6
0
0
0
0
0
0
0
0
2
2
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
11
0
0
0
0
0
3
0
0
0
0
0
0
0
0
0
1 0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
6
12
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
0
0
15
0
0
1
0
0
1
0
0
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
1
0
0
0
1
0
1
0
0
11
3
0
0
0
0
0
0
2
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
12
1
1
0
1
0
0
0
0
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
16
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
1
0
0
0
2
15
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
17
0
0
0
0
0
0
0
0
0
2
1
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
1
0
0
0
0
0
14
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
1
0
0
15
0
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
0
0
0
0
1
0
1
1
0
13
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
2
1
1
0
0
1
0
11
0
0
0
0
0
0
0
0
1
0
0
1
1
5
1
0
1
0
1
0
0
0
0
0
0
0
0
1
0
0
0
0
0
8
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
1
1
0
0
1
2
1
0
0
0
0
0
0
12
Table 6-Confusion Matrix of Discrete Wavelets using KNN with Database (c)
-
-
Conclusion
The system discussed here consists of three types of feature extraction methods, three types of classifiers and three types of sets of databases.Comparative analysis of recognition rate using thee three different combinations shows that:
-
If Training samples are more than testing samples than recognition rate is high.
-
If Discrete Wavelets are used as Features i.e Discrete Wavelet Transform is used as Feature Extraction method than recognition rate is comparatively more.
-
If K-Nearest Neighbor is used as Classifier than recognition rate is high.
References
-
Yuk Ying Chung, Man To Wong; Handwritten Character Recognition By Fourier Descriptors And Neural Network; IEEE TENCON on Speech and Image Technologies for Computing and Telecommunications, 1997, Australia, page(s):391- 394.
-
Anshul Gupta, Manisha Srivastava, Chitralekha Mahanta; Offline Handwritten Character Recognition Using Neural Network; IEEE International Conference on Computer Applications And Industrial Electronics(ICCAIE), 4-7 Dec. 2011, Penang, page(s):102-107.
-
Bhattachraya U., Vajda S., Mallick A., Chaudhuri B.B., Belaid A., On The Choice Of Training Set, Architecture and Combination Rule Of Multiple MLP Classifiers For Multiresolution Recognition Of Handwritten Characters IEEE Transactions On frontiers in Handwriting Recognition, Kolkata, India, page(s):419-424, 2004.
-
S.B Patil, G.R Sinha, V.S Patil, Recognition of Devnagri Chracters 2nd International Conference on emerging Applications of Information technology (EAIT), India, page(s):185-189, 2011.
-
Arora S., Bhattacharjee D., Nasipuri M.; Combining Mulitiple Feature Extraction Techniques for Handwritten Devnagri Character Recognition; IEEE Region 10 and the Third International Conference on Industrial and Information Systems, 8-10 Dec. 2008, Kharagpur, page(s):1-6.
-
Rafael C. Gonzalez, Richard E. Woods, Steven L. Eddins, Digital Image Proceesing Using
MATLAB, publication: Dorling Kindersely (India) Private Limited, edition:2006.
-
Database: ISI, Kolkata