IDENTIFICATION OF DEVELOPMENTAL
DYSGRAPHIA BY HANDWRITING
ANANLYSIS
By
Abhimanshu Kumar
ShaluKumari
Satyam Garodia
Richa Sinha
UNDER THE GUIDANCE OF
Mr. Rajib Saha
PROJECT REPORT SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
BACHELOR OF TECHNOLOGY IN COMPUTER SCIENCE AND
ENGINEERING
RCC INSTITUTE OF INFORMATION TECHNOLOGY
Session 2017-2018
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
RCC INSTITUTE OF INFORMATION TECHNOLOGY
[Affiliated to West Bengal University of Technology]
CANAL SOUTH ROAD, BELIAGHATA, KOLKATA-700015
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
RCC INSTITUTE OF INFORMATION TECHNOLOGY
TO WHOM IT MAY CONCERN
I hereby recommend that the Project entitled IDENTIFICATION OF
DEVELOPMENTAL DYSGRAPHIABY HANDWRITING ANANLYSIS
preparedunder my supervision by Abhimanshu Kumar (Reg. No. 141170110001,
Class Roll No. CSE2014/063),Richa Sinha (Reg. No. 141170110048, Class Roll
No. CSE2014/083), Satyam Garodia (Reg. No. 141170110062, Class Roll No.
CSE2014/080), ShaluKumari (Reg. No. 141170110064, Class Roll No.
CSE2014/064) of B.Tech (8
th
Semester), may be accepted in partial fulfillment for
the degree of Bachelorof Technology
inComputer Science& Engineering under
Technology(WBUT).
.
…………………………………………
Project Supervisor
Department of Computer Science and Engineering
RCC Institute of Information Technology
Countersigned:
………………………………………
Head
Department of Computer Sc. &Engg,
RCC Institute of Information Technology
Kolkata – 700015.
West Bengal University of
DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING
RCC INSTITUTE OF INFORMATION TECHNOLOGY
CERTIFICATE OF APPROVAL
The foregoing Project is hereby accepted as a credible study of an
engineering subject carried out and presented in a manner satisfactory to warrant
its acceptance as a prerequisite to the degree for which it has been submitted. It
is understood that by this approval the undersigned do not necessarily endorse or
approve any statement made, opinion expressed or conclusion drawn therein, but
approve the project only for the purpose for which it is submitted.
FINAL EXAMINATION FOR 1. —————————————
EVALUATION OF PROJECT
2. ———————————
(Signature of Examiners)
ACKNOWLEDGEMENT
The satisfaction that accompanies the progress of this work would be
incomplete without the mention of the people who made it possible, without whose
constant guidance and encouragement would have made efforts go in vain. I
consider myself privileged to express gratitude and respect towards all those who
guided us through the considerable progress of this project.
We convey thanks to our guide Mr. RAJIB SAHA of Computer Science and
Engineering Department for providing encouragement, constant support and
guidance which was of a great help and motivation to us.
————————————
————————————
————————————
————————————
Table of Contents
Page No.
1. Introduction ……………………………………………………. 6
2. Review of Literature …………………………………………… 8
3. Objective of the Project………………………………………… 13
4. System Design…………………………………………………… 14
5. Methodology for implementation (Formulation/Algorithm)…… 15
6. Implementation Details…………………………………………. 16
7. Results/Sample output………………………………………….. 26
8. Conclusion………………………………………………………. 36
References
1. Introduction
Dysgraphia is a specific learning disability that affects written expression.
Dysgraphia can appear as difficulties with spelling, poor handwriting and trouble putting
thoughts on paper. Dysgraphia can be a language based, and/or non-language based disorder.
Many people have poor handwriting, but dysgraphia is more serious. Dysgraphia is a
neurological disorder that generally appears when children are first learning to write. Experts
are not sure what causes it, but early treatment can help prevent or reduce problems.
Writing requires a complex set of motor and information processing skills. Not only does it
require the ability to organize and express ideas in the mind. It also requires the ability to get
the muscles in the hands and fingers to form those ideas, letter by letter, on paper.
Dysgraphia that is caused by a language disorder may be characterised by the person having
difficulty converting the sounds of language into written form or knowing which alternate
spelling to use for each sound. A person with dysgraphia may write their letters in reverse,
have trouble recalling how letters are formed, or when to use lower or upper case letters. A
person with dysgraphia may struggle to form written sentences with correct grammar and
punctuation, with common problems including ommitting words, words ordered incorrectly,
incorrect verb and pronoun usage and word ending errors. People with dysgraphia may
speak more easily and fluently than they write.
SOME SIGNS OF DYSGRAPHIA
o Generally illegible writing
o Inconsistencies in writing, e.g. mixtures of printing and cursive writing,etc
o Unfinished words or letters, omitted words
o Inconsistent position of letters on the page with respect to lines and margins
o Inconsistent spaces between words and letters
o Strange wrist, body, or paper position
o Talking to self whilst writing, or carefully watching the hand that is writing
o Slow or laboured copying or writing
o Large gap between written ideas and understanding demonstrated through speech.
o Difficulty organising thoughts on paper
CLASSIFICATION
Dysgraphia is nearly always accompanied by other learning differences such
as dyslexia or attention deficit disorder, and this can impact the type of dysgraphia a person
might have. There are three principal subtypes of dysgraphia that are recognized. There is
little information available about different types of dysgraphia and there are likely more
subtypes than the ones listed below. Some children may have a combination of two or more
of these, and individual symptoms may vary in presentation from what is described here.
Most common presentation is a motor dysgraphia resulting from damage to some part of the
motor cortex in the parietal lobes.
Dyslexic
People with dyslexic dysgraphia have illegible spontaneously written work. Their copied
work is fairly good, but their spelling is usually poor. Their finger tapping speed (a method
for identifying fine motor problems) is normal, indicating that the deficit does not likely
stem from cerebellar damage.
Motor
Motor dysgraphia is due to deficient fine motor skills, poor dexterity, poor muscle tone, or
unspecified motor clumsiness. Letter formation may be acceptable in very short samples of
writing, but this requires extreme effort and an unreasonable amount of time to accomplish,
and it cannot be sustained for a significant length of time, as it can cause arthritis-like
tensing of the hand. Overall, their written work is poor to illegible even if copied by sight
from another document, and drawing is difficult. Oral spelling for these individuals is
normal, and their finger tapping speed is below normal. This shows that there are problems
within the fine motor skills of these individuals. People with developmental coordination
disorder may be dysgraphia. Writing is often slanted due to holding a pen or pencil
incorrectly.
[2]
Spatial
A person with spatial dysgraphia has a defect in the understanding of space. They will have
illegible spontaneously written work, illegible copied work, and problems with drawing
abilities. They have normal spelling and normal finger tapping speed, suggesting that this
subtype is not fine motor based.
2. Review of Literature
Handwriting, which is a required activity among school-aged children, involves both
spatial and temporal demands (Amundson & Weil, 1996; Tseng & Chow, 2000).
Handwriting performance is considered to be proficient when legible text is produced at a minimum
of effort. In this case, handwriting is automatic and does not interfere with the content as generated
by the creative thinking process (Scardamailia, Bereiter, & Goleman, 1982). In contrast, poor
handwriters are often unable to achieve a completely automated process, and their handwriting may
be slow and unclear.Handwriting difficulty, or dysgraphia, was defined by Hamstra-Bletz and Blote
(1993) as a disturbance or difficulty in the production of written language that is related to the
mechanics of writing. Teachers have estimated that 11–12% of female students and 21–32% of male
students have handwriting difficulties (Rubin & Henderson, 1982; Smits-Engelsman, Van Galen,
&Michels, 1995).
Two main outcomes have been used to assess and define poor handwriting, namely,product
readability or legibility and performance time. Product legibility has been evaluated in two ways:
(1) by judging the readability of an entire paragraph (Ayres, 1912; Freeman,1959),
Or
(2) by analytic methods based on grading specific features that characterize
readability (e.g., between letter and word spacing, letter formation, the degree of line slant,etc.) and
then calculating an overall score (see Rosenblum, Weiss, &Parush, 2003, for more details).
There are a number of reasons why current handwriting assessments are of limited
value. First, their reliability is low to moderate; second, they require prolonged processing time by
the evaluator who needs to judge the writing product for each of the legibility criteria and third, they
do not provide substantive information about the writing process(Rosenblum et al., 2003,
Rosenblum, Weiss, &Parush, 2004). The third reason poses a significant limitation, as it is believed
that only a comprehensive description of the realtime, dynamic characteristics of a child‟s
handwriting can provide insight into the motor control mechanisms of normal handwriting and an
understanding of the underlying pathology of handwriting difficulties (Dobbie & Askov, 1995;
Graham & Weintraub, 1996;LongstaV& Heath, 1997; Sovik, Arntzen, &Thygesen, 1987a, 1987b).
In recent years, more attention has been devoted to identifying the features of poor
handwriting by children who have a variety of perceptual-motor and learning problems
(e.g., Rosenblum, Parush, & Weiss, 2001; Rosenblum et al., 2003; Schoemaker& Smits-Engelsman,
1997; Smits-Engelsman, Van Galen, &Portier, 1994a; Smits-Engelsman,Niemeijer, & Van Galen,
2001). In most of these studies, children were asked to perform brief writing tasks (i.e., usually a
single sentence). The testing of only brief writing tasks has limited ecological validity, since many
clinicians and educators, as well as researchers, indicate that handwriting problems are particularly
noticeable during the performance of tasks similar to those occurring in thechildren‟s natural learning
environment (Rosenblum et al.,2003, 2004).
The term Dysgraphia is not widely used in schools. One reason is that handwriting difficulties can be
included under the label of learning disabilities. Another reason is that there is no consensus in the
field on one definition or identification process for dysgraphia. Richards (1999) defines dysgraphia
as a problem with expressing thoughts in a written form. Meese (2001) describes dysgraphia as
handwriting problems, specifically, a partial inability to remember how to make certain alphabet or
arithmetic symbols. For the purpose of this paper, we are using the latter definition, and will be
focusing on the mechanics of handwriting.
The treatment of dysgraphia can be elusive. Many instructional strategies have been
proposed to help students with dysgraphia, but only some have empirical evidence to support them.
Teachers should be aware of the signs and symptoms of dysgraphia and not dismiss a child as
simply having sloppy handwriting. If a teacher starts to see a trend of illegible writing, it is
appropriate for them to question whether this child has dysgraphia. Teachers should note which parts
of the writing process are most difficult for the child. While dysgraphia often occurs along with
another dis-ability, many students with dysgraphia can exhibit high academic achievements in other
subjects (Richards, 1999). Figure 1 shows an example of the handwriting of a second grade student
with dysgraphia, and a typical second grader‟s handwriting. The characteristics of dysgraphia are
varied and students can exhibit any one or more of these characteristics (see box, “Characteristics of
Dysgraphia”).
Feifer (2001) believes that dysgraphia can be categorized into four subtypes. The first subtype is
phonological dysgraphia, that is “writing and spelling disturbances in which the spelling of
unfamiliar words, nonwords, and phonetically irregular words are impaired” (p. 1). These students
tend to have trouble spelling by sounds and rely on the visual aspect of letters; therefore, because
spelling is an auditory task, they will have trouble with spelling tests. The second sub-type is surface
dysgraphia where students have trouble with orthographic representations of words, which makes the
student rely too heavily on sound patterns; the opposite of phonological dysgraphia. Mixed
dysgraphia is the third subtype of dysgraphia. This type refers to students having trouble with mixing
up letter formations and having trouble with spelling tasks, a combination of the first two types.
Recalling letter formations is hard for these students to do because there are so many instructions or
rules that they get con-fused and; therefore, have inconsistent spellings of words. Finally,
semantic/syntactic dysgraphia is a grammatical problem in which students have difficulty with how
words can be joined to make complete and comprehensive phrases.
In addition, children with dysgraphia usually have some type of problem with automaticity that
interferes with the retrieval of letter formation (Richards, 1999). The con-centration on how to form
the letter overwhelms the child to a degree that the letter is written poorly. Incorrect letter or word
formation can also lead to exceeding the margins or lines. Letter formation is automatic for most
students after initial skill attainment. When letter formation is automatic, students can concentrate on
spelling, grammar, sentence structure, and other aspects of written language. However, for many
students with dysgraphia, letter formation is a cognitive task which leaves little mental capacity to
devote to these other aspects. Children with dysgraphia can become frustrated, leading to low
motivation to use and practice written language.
Students concentrating too hard on letter formation may develop problems with gripping the pencil
(Richards, 1999). A list of characteristics of a poor pencil grip can be seen in the box below.
Gripping the pencil a “wrong” way can interfere with performance because the child focuses on
holding the pencil instead of writing the letter. Richards (1998) suggested a proper pencil grip that
included placing the fingers about one inchabove the tip of the pencil, maintaining a 45 degree angle
with the paper, and using moderate pressure. Teachers should be aware of a child holding the pencil
in an improper way and aim to correct the grip.
In addition to pencil grip and automaticity, retaining information in the working memory is not
something most students have trouble mastering. Children with dysgraphia, however, often exhibit
trouble with working memory because so much of their cognitive energy is put into the mechanical
level of writing letters or words (Richards, 1999). It is similar to the seven plus or minus two phe-
nomenon; the hypothesis that claims one can only hold seven items in memory plus or mi-nus two.
For example, a social security number is nine items long. However, by clustering the nine digits into
three groups, most people can easily recall their social security number. Children already have a
limited number of spaces to hold information in memory and therefore have a harder time
remembering many pieces of information. This study did not support the opinion that dysgraphia is a
temporary developmental delayfor children. While this study did not involve an intervention, the
researchers theorized that dysgraphia may be sensitive to training that addresses better movement
strategies
One such remedial treatment is using drill and practice. Ediger (2002) suggested that the
teacher should provide a clear example of good handwriting and then the children should practice
and drill using the teacher‟s model. People with dysgraphia struggle with the display of letters
because often the letter that is asked for in the brain is not the letter that is retrieved and produced
(Richards, 1998). Repetitive practice, along with correct position and pencil grip can help with this
process.
Another remedial treatment that has empirical evidence is building fine motor skills. Using drills that
build the muscles used for fine motor activities can help improve hand functioning, which can lead to
better handwriting (Berry, 1999). Keller (2001) used such activities in a club she created to help the
handwriting of students with dysgraphia. Dikowski (1994) studied children‟s visual-motor skills
related to handwriting. He found schools offered little help to students with handwriting or visual-
motor disabilities. He observed that when children had visual-motor integration problems this led to
problems with hand-eye coordination. Since the brain, hand, and eye all work together to per-form
anything written.
IMAGE PROCESSING:-
Sight is a human being‟s principle sense. A visual image is rich in information from the outer world
and receiving and analyzing such images is part of the routine activity of human
beings throughout their walking lives. At a more sophisticated level, human beings may generate
record or transmit images. These activities together comprise image processing.
Theories and techniques of image processing originated in the study of optics and optical
instruments. However, the advent of digital computers opened vast new possibilities
for artificial image processing. By the mid-1960‟s, third-generation computers offered the speed and
s to r age necessary for practical
implementation of image-processing algorithms; and in 1964 the capabilities of digital image
processing were spectacularly demonstrated when pictures of the moon transmitted by the Ranger 7
space probe were processed to correct various types of image distortion inherent in the on-board
television camera .Since that date, the field of image processing has experienced vigorous growth.
Digital image processing techniques are used today in a wide range of applications that,
although otherwise unrelated, share a common need for methods capable of enhancing pictorial
information for huma n interpretation and analysis .These applications include: remote sensing;
security monitor in g;medical diagnosis ; automatic inspection; radar; sonar; detection of military
targets; robotics; business communication; television enhancement ; etc.
Graphology:
Graphologyis the analysis of the physical characteristics and patterns of handwriting purporting to be
able to identify the writer, indicating psychological state at the time of writing, or evaluating
personality.
The word Graphology is derived from grapho- (from the Greek γραφή, "writing") and logos (from
the Greek λόγος, which relates to discussion or
theory).
Graphology is a blend of art and science. It is a science because it measures the structure and
movement of the written forms – slants,angles and spacing are accurately calculated and the pressure
is observed in magnification and with precision. And it is an art because the graphologisthas to
constantly keep in mind the total context in which the writing is
taking place: the gestalt of the writing as a whole.
Graphotherapy:
Graphotherapy is a form of reverse psychology accomplished through handwriting exercises
specifically designed to bring about more desirable personality traits. By doing these exercises you
are accomplishing a gradual regrooving or retraining of your subconscious mind.
Graphotherapy works with the subconscious to make changes to those personality traits. You may
think how you write is a conscious choice, after all when you choose to do your best writing, it looks
different from the quick scrawl you make as you take notes from a phone call.
How does graphotherapy work?
Each stroke you make means something. It connects to a personality trait, a habit, a way of being that
is presently part of your personality make-up. Sometimes there is more than one stroke that indicates
the same or very similar tendencies. To add all the ones you want and replace all the ones you don„t
want is the goal of graphotherapy.
APPLICATIONS AND LIMITATIONS OF GRAPHOLOGY:-
Applications of Graphology:
There are many uses of handwriting analysis. Below are a few of the most popular applications we
use today. They are:
• Dating and Socializing
• Employee hiring and human resources
• Police profiling
• Self improvement and professional speaking
• Councillor, therapist and coaching applications
LIMITATIONS OF HANDWRITING ANALYSIS:-
Below are a few of the limitations of Handwriting analysis:
reveal the AGE of the writer
the right hand or any other part of the body.
cannot be found in
the handwriting.
LIMITATIONS WHEN WE DO ANALYSIS USING IMAGE PROCESSING TECHNIQUES
ARE:
Below are a few of the limitations when we do analysis using image processing technique:
cannot know the pressure of writing.
IMAGE SEGMENTATION:
into different components (thus to facilitate the task at higher levels such as object
detection and recognition)
-level vision task
performed by neurons between low-level and high-level cortical areas
In Handwriting image segmentation digital handwriting is segmented into three
different types of segments, i.e. word segmentation, letter segmentation and line
segmentation, each used for different processing.
The need to have clear, neat handwriting is of utmost importance in today‟s society.
Communicating ideas, writing and signing checks, signing legal agreements, and other daily
activities need clear handwriting that is legible by others. One may argue that technology can replace
the need for handwriting. For example, paying bills is now available online. However, computers
cannot be relied on for everything. One factor to consider is the technology gap. There are many
homes and work places that do not have computers and there are many other instances in daily life
when a computer is not readily available. ! Furthermore, as young children learn the writing process
and how to formulate thoughts in writing, the use of technology may not be practical. The physical
act of writing down one‟s thoughts is part of the cognitive process of learning to communicate
through writing. A young child who has not yet learned these skills would not be able to transfer the
skills to a word processing program. The outcome of this study provided evidence that using drill and
fine motor ac-tivities together greatly improved the hand-writing of a second grader with dysgraphia.
While Sam showed improvement over the eight weeks during the after school sessions, he had a hard
time generalizing what he learned to his class work. His written work improved, but Sam continued
to make a few letters the same way he did before the inter-vention. Over time, it is my hope that he
will continue to generalize and remember how to write each letter. There were some limitations to
this action research. One limitation was that the study used only one participant. The purpose of
action research is to identify a problem within a classroom and address the problemstudy met this
purpose of action research, the findings are not generalizable. However, other teachers can learn
from this case study both in terms of knowledge regarding dysgraphia and options for intervention.
Teachers who work with children struggling with handwriting can gain information and tech-niques
to help guide handwriting remediation, even if the child is not diagnosed with a writ-ing disability.
Students in all elementary grades could benefit from structured instruc-tion on handwriting and how
to form letters.
In 2016, Abhishek Bal and RajibSaha
[1]
proposed an off-line handwritten document analysis through
segmentation, skew recognition and writing pressure for cursive handwritten document. The
proposed segmentation method is based on modified horizontal and vertical projection that can
segment the text lines and words even if the presence of overlapped and multi-skewed text lines.
Proposed work also present orthogonal projection based baseline recognition and normalization
method as well as writing pressure detection method that can predict the personality of writer from
the baseline and writing pressure.
3. Objective of the project
Dysgraphia is one of the major problems found in children. Though it is short term effect but if it
goes untreated remains throughout life.
So, through this project we have tried to find out the similarity between the people suffering from
this disease. Also to bring out the matching pattern and the personality characteristics of the
personality.
The hand writing samples are collected from wide variety of people belonging to both economical
stable family and also from weaker section of the society.
Samples were also collected from equal number of males and females.
The purpose is to chalk out the characteristics of the people suffering from dysgraphia and
characteristics related to the disease.
4.System Design
SOFTWARES USED:
We have mainly used Microsoft Windows 8 as the platform for implementing the
Handwriting Recognition System. Also we have used MATLAB as the coding platform
to implement the system and depict the procedure of each step and display the
corresponding output. The software requirements in the system building the process are
as follows
Microsoft Windows 8:
MATLAB:
HARDWARE USED:
The following hardware components needed in combination to our system for successful
handwriting analysis are described below:
Webcam/Digital Camera
4 GB RAM
Keyboard
Mouse
Monitor
5.Methodology for implementation(formulations/Algorithm)
5.1 Algorithm for the system
Step 1. Take samples input collected
Step 2. Convert the image into gray image
Step 3. Convert the image into binary image
Step 4. Getting the size of the image
Step 5. Segment the image into lines words characters
Step 6. Compare the characteristics properties
Step 7. Suggesting the result.
5.2 Algorithm for line segmentation
Step 1. Read the images iteratively
Step 2. Convert the image into gray image
Step 3. Convert the image into binary image
Step 4. Count and sum the pixel values row-wise fashion & store in 1D array
Step 5. Traverse the 1D array containing row-wise pixel count
Step 5.1. if pixel count > 9 then
Step 5.1.2 do till pixel count <9 , segment the region & display
Step 5.1.2.1 Store that segmented part in a folder
Step 6. End
5.3 Algorithm for word segmentation
Step 1. Read the segmented lines image iteratively
Step 2. Set pixel= 255* no. Of rows
Step 3. Count and sum the pixel values column-wise fashion & store in 1D Step 4.
Traverse from 1
st
column to last
Step 5. Traverse the 1D array containing column-wise pixel count
Step 5.1. if pixel count not equal pixel then
Step 5.1.2 do till pixel count = pixel, segment the word
Step 5.1.2.1 Store that segmented part in a folder
Step 6. End
5.4 Algorithm for letter segmentation
Step 1. Read the segmented words image iteratively
Step 2. Set pixel= 255* (r-1)
Step 3. Count and sum the pixel values column-wise fashion & store in 1D Step 4.
Traverse from 1
st
column to last
Step 5. Traverse the 1D array containing column-wise pixel count
Step 5.1. if pixel count not equal pixel then
Step 5.1.2 do till pixel count = pixel, segment the letter
Step 5.1.2.1 Store that segmented part in a folder
Step 6. End
5.5 Algorithm for template matching
Step 1. Read the template image
Step 2. Read the target image
Step 3. Convert both the images into gray scale if they are in rgb format
Step 4. Find the mean of the template image
Step 5. Find the mean of the target image
Step 6. Subtract mean value from the template image matrix
Step 7. Subtract mean value from the target image matrix
Step 8. Find covariance between template and target image
Step 9. Co-relate the images and box the matching result
6.Implementation details
• Handwriting samples were collected from 50 people
• 30 were children of age group 6-15 years and 20 were of 15+ age
• Out of this 50, 25 were males and rest females
• We will use a digitizing tablet to acquire handwriting
• employ complex parameterization to quantify its kinematic aspects and
hidden complexities.
• begins with collecting the handwriting samples on plain white A4 size paper
• perform preprocessing steps such as binarization and noise removal etc for better
recognition
• color image or gray scale image is taken as an input then thresholding is done to convert the
image into binary image
1. To remove the noise from the image
2. To get the better results
3. For accuracy and correction purposes
• Then line segmentation, word segmentation and character segmentation have been
performed
• After each segmentation process, the characteristics properties are matched.
6.1 Sample Collection
Handwriting samples of 50 children were collected and it was found that out of these 50 , 4 person
were actually suffering from dysgraphia. This samples were taken on A4 sheets and considered
further.
6.2 Image Pre-processing
Image pre-processing [1] is the technique in which the handwritten sample is translated into a
format which can be easily and efficiently processed in further steps. These steps involve
binarization, noise removal, line segmentation, word segmentation and skew normalization.
Binarization converts gray scale image into binary image. Quality of the converted image is
improved by applying noise removal techniques.
6.3 Line Segmentation
After binarization and noise removal, the converted image is processed through a line segmentation
[1]
technique to split the hand-written document into separate lines based on rising section of the
horizontal projection histogram of document image.
6.4 Word Segmentation
In the case of word segmentation [1], to segment, the words from the line, firstly inter-word and
intra-word gaps are measured. Inter-word gaps denote the gaps between two words and intra-word
gaps denote gaps within a word. Generally, gaps between the words are larger than the gaps within a
word. These proposed methods construct the vertical projection histogram to measure the width of
each inter-word and intra-word gaps then it measures the threshold value to differentiate between
inter-word and intra-word gaps. If the width of gaps is greater than or equals to threshold then gaps
are treated as inter-word gaps and words are segmented individually from the line depending on the
threshold.If a line has global skew then it may possible that several words within a line may have
different skew. So, it may require normalizing the skew of the words for a single line. For that
reason, again skew normalization method is applied to each segmented word separately.
6.5Code for line segmentation
clc
clear
files=dir('G:\matlab\work\final\images\*.jpg');
line=1;
img1=zeros(1000,1000);
while (line<=length(files))
filename=strcat('G:\matlab\work\final\images\',files(line
).name);
a=imread(filename);
g= rgb2gray(a);
minimum=min(g(:));
maximum=max(g(:));
med=(minimum+maximum)/2;
[rows, col]=size(g);
fori=1:rows
for j=1:col
if g(i,j)>= med
img1(i,j)=1;
else
img1(i,j)=0;
end
end
end
imshow(img1);
% pause
p=zeros(rows,1);
fori=1:rows
p(i,1)=col - sum(img1(i,:));
end
fori=1:rows
if(p(i,1)>9)
for j=i+1:rows
if(p(j,1)<9)
break;
end
end
imshow(img1(i:j-1,:));
end
end
end
6.6 code for word segmentation
files=dir('G:\matlab\work\final\lines\*.jpg');
word_no=1;
line=1;
while (line<=length(files))
filename=strcat('G:\matlab\work\final\lines\',files(line)
.name);
z1=imread(filename);
%figure,
%imshow(z1);
[r, c]=size(z1);
pixel=255*r;
q=zeros(c,1);
for j=1:c
q(j,1)= sum(z1(1:r,j));
end
f=1;
while(f<=c)
if(q(f,1)~=pixel)
for g=f:c
if (q(g,1)==pixel)
break;
end
end
word_img=(z1(:,f:g-1));
figure,
imshow(word_img);
filename=sprintf('word%d.jpg',word_no);
path=strcat('G:\matlab\work\final\words\',filename);
imwrite(word_img,path);
f=g-1;
word_no=word_no+1;
end
f=f+1;
end
line=line+1;
end
6.7 code for letter segmentation
files=dir('G:\matlab\work\final\chars\*.jpg');
no=1;
line=1;
while (line<=length(files))
filename=strcat('G:\matlab\work\final\chars\',files(line)
.name);
a=imread(filename);
%a=imread('G:\matlab\work\char_test\word10.jpg');
%imshow(a)
[r,c]=size(a);
cs=zeros(c,1);
fori=1:c
cs(i,1)=sum(a(:,i));
end
i=1;
while(i<=c)
if(cs(i,1)<8350)
for j=i+1:c
if(cs(j,1)>8350)
break;
end
end
imgs=(a(:,i:j-1));
figure,
imshow(imgs)
filename=sprintf('char%d.jpg',no);
path=strcat('G:\matlab\work\final\chars\',filename);
imwrite(imgs,path);
no=no+1;
i=j-1; %increment the value to starting of the next
line
end
i=i+1;
end
line=line+1;
end
6.8 code for template matching
6.8.1 main
im1=imread('G:\matlab\work\final\char\word92.jpg');
im2=imread('G:\matlab\work\final\char\wordd14.jpg');
result1=tmp(im1,im2);
figure,
subplot(2,2,1),imshow(im1);title('Template');
subplot(2,2,2),imshow(im2);title('Target');
subplot(2,2,3),imshow(result1);title('Matching Result
within black box');
6.8.2 template-target matching
function result=tmc(image1,image2)
if size(image1,3)==3
image1=rgb2gray(image1);
end
if size(image2,3)==3
image2=rgb2gray(image2);
end
% check which one is target and which one is templete
if size(image1)>size(image2)
Target=image1;
Template=image2;
else
Target=image2;
Template=image1;
end
% read both images sizes
[r1,c1]=size(Target);
[r2,c2]=size(Template);
% mean of the template
image22=Template-mean(mean(Template));
%corrolate both images
corrMat=[];
fori=1:(r1-r2+1)
for j=1:(c1-c2+1)
Nimage=Target(i:i+r2-1,j:j+c2-1);
Nimage=Nimage-mean(mean(Nimage)); % mean of image part
under mask
corr=sum(sum(Nimage.*image22));
corrMat(i,j)=corr;
end
end
% plot box on the target image
result=plotbox(Target,Template,corrMat);
6.8.3 plotting box
function result=plotbox(Target,Template,M);
[r1,c1]=size(Target);
[r2,c2]=size(Template);
[r,c]=max(M);
[r3,c3]=max(max(M));
i=c(c3);
j=c3;
result=Target;
for x=i:i+r2-1
for y=j
result(x,y)=00;
end
end
for x=i:i+r2-1
for y=j+c2-1
result(x,y)=00;
end
end
for x=i
for y=j:j+c2-1
result(x,y)=000;
end
end
for x=i+r2-1
for y=j:j+c2-1
result(x,y)=000;
end
end
7.Sample output and result
7.1 Segmentated Lines sample
1.
2.
3.
4.
5.
6.
7.
7.2 segmentated word samples
7.3 letter segmentation result
7.4 template matching results
Unmatched data
Results Observed
Letter
Observed style
Imposed characterstics
e
Floating tail, curved head
Uninfluenced by emotions
i
Left slanted
Reserved & introspect
G
Squeezed and curved
Selective in nature
S
backward tilted, narrow
bottom
Holding aggression
.
8. Conclusion
So, this results signify that people with Dysgraphia or tending to get dysgraphia will have the above mentioned
characteristics with the similarity in the pattern. But it is does not in any case tell that a person not having
dysgraphia will not show this properties.
Of the 50 peoples sample collected, 4 people showed signs of dysgraphia. Among this 4, 3 were children
belonging to age group of 6-15 years.
The results also show that dysgraphia has no connection with the gender of that person. It can happen to both
male and female.
Also, of this 4 people, 3 belonged to the economically stable family and the other person was from lower
middle class family.
So we can say that dysgraphia has nothing to do with the economic status of the family from which the person
belong.
References
Handwriting evaluation for developmental dysgraphia :Process versus product
1. Sara Rosenblum, Ph.D, Patrice L. Weiss, Ph.D and Shula Parush, Ph.D
2. Automatic segmentation as a tool for examining the handwriting process of children with
dysgraphic and proficient handwriting
Sara Rosenblum ,Assaf Y. Dvorkin , Patrice L. Weiss
Dikowski, T. J. (1994). Educational interven-tions for visual-motor deficiencies that affect
handwriting in school-aged children. Ed. D. Practicum Report,Nova Southeastern
University. (ERIC Document Reproduction Service No. ED 74 02)
Ediger, M. (2002). Assessing handwriting achievement. Reading Improvement,
39, 103-110.
Feifer, S. G. (2001) Subtypes of language-based dysgraphias. Handout fromWorkshop “Learning Of
The Brain: Using Brain Research To Leave No Child Behind” at the Hyatt Regency. Boston,
MA: Public Information Re-sources.
Graham, S., & Harris, K. R. (1988). Instruc-tional recommendations for teaching writing to
exceptional students. Ex-ceptional Children, 54, 506-512.
Graham, S., Harris, K. R., & Fink, B. (2000). Is handwriting causally related to learning to writing?
Treatment of handwriting problems in beginning writers. Journal of Educational Psy-
chology, 92, 620-633.
