IDENTIFICATION OF

DEVELOPMENTAL DYSGRAPHIA
BY HANDWRITING ANANLYSIS
By
Subhajit Bose
Ipsita Chakraborty
Nabanita Halder

UNDER THE GUIDANCE OF

Ms. Monika Singh

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 DYSGRAPHIA BY HANDWRITING ANANLYSIS
prepared under my supervision by Subhajit Bose (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), Shalu Kumari (Reg. No. 141170110064, Class Roll No.
CSE2014/064) of B.Tech (2nd Semester), may be accepted in partial fulfillment for

West Bengal University of

the degree of Bachelor of Technology in Computer 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.
 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 …………………………………………………….
2. Review of Literature ……………………………………………
3. Objective of the Project…………………………………………
4. System Design……………………………………………………
5. Methodology for implementation (Formulation/Algorithm)……
6. Conclusion……………………………………………………….
References
1. Introduction

Developmental dysgraphia, being observed among 10-30% of school-aged children, is

a disturbance or difficulty in the production of written language that has to do with the
mechanics of writing. The objective of this study is to propose a method that can be
used for automated diagnosis of this disorder, as well as for estimation of difficulty level
as determined by the handwriting proficiency screening questionnaire.
We used a digitizing tablet to acquire handwriting and consequently employed a
complex parameterization in order to quantify its kinematic aspects and hidden
complexities. We also introduced a simple intra-writer normalization that increased
dysgraphia discrimination and HPSQ estimation accuracies. Using a random forest
classifier, we reached 96% sensitivity and specificity, while in the case of automated
rating by the HPSQ total score, we reached 10% estimation error.
This study proves that digital parameterization of pressure and altitude/tilt patterns in
children with dysgraphia can be used for preliminary diagnosis of this writing disorder.

Dysgraphia:

• Transcription disability - writing disorder

• impaired handwriting, orthographic coding & finger sequencing
Classification of Dysgraphia

1. Dyslexic
• Illegible spontaneously written work.
• Their copied work is fairly good
• But spelling is usually poor

2. Spatial
• defect in the understanding of space
• illegible spontaneously written work
• illegible copied work
• problems with drawing abilities
3. Motor
• deficient fine motor skills
• poor muscle tone

2. Review of Literature

“Learning disability is a general term that refers to a heterogeneous group of disorders

manifested by significant difficulty in the acquisition and use of listening, speaking, reading,
writing, reasoning, or mathematical abilities. These disorders are intrinsic to the individual,
presumed to be due to central nervous system dysfunction, and may occur across the life
span” (NJCLD, 1981). “A learning disability refers to a retardation, disorder, or delayed
development in one or more of the processes of speech, language, reading, writing,
arithmetic, or other school subject resulting from a psychological handicap caused by a
possible cerebral dysfunction and/or emotional or behavioural disturbances. It is not the
result of mental retardation, sensory deprivation, or cultural and instructional factors.”
(Kirk, 1962, Educating Exceptional Children, p. 263) The definition of Learning Disabilities in
the Federal Law (IDEA-2004) defined the term specific learning disability means a disorder
in one or more of basic psychological process involved in understanding or in using
language spoken or written, which disorder may manifest it self in imperfect ability to listen,
think, speak, read, write, spell or to do mathematical calculations. Carlson, Shirley (2005)
depicted a two hundred year history of learning difficulties. There is a vast difference
between a learning difficulty and a learning Disability. An individual with learning difficulty
can learn using conventional teaching techniques while learning disability (LD) requires
specialized interventions which depend on the type of disability. The most common forms
of LD are dyslexia also known as word blindness, dyscalculia, dysgraphia, and short term
memory dysfunction. LD can result from injury, it can be hereditary it can come in many
forms. Although it is no longer considered a form of mental retardation, it is now known
that true LD is directly related to some type of brain malfunction. They found that LD did
not originate in the 1960s as some people believe but actually spans the time from
Napoleonic reign to space age exploration. Scientists, researchers, and educators know
quite a bit about LD now, but they still do not know nearly enough to help the multitude of
individuals afflicted with those conditions. John Elkins (2007), conducted a study on
“Learning Disabilities”. He advocated an approach to support students who experience
difficulties in learning particularly in the key areas of literacy and numeracy. In the state of
Queensland, Australia, a distinction has been made between students experiencing learning
difficulties and those who have learning disabilities. However government priorities for
improved achievement in literacy and numeracy made schools focus on the performance of
all low achieving students without regard to diagnostic category. Throughout this article the
different yet convergent understandings of LD in Australia and the United States are tracked
with suggestions made for the future research that avoid the problems of operational the
definition of LD proposed by Keogh (1982). Diagnostic and Statistical Manual of Mental
Disorders (DSM)’s definition of Specific Learning Disorder (SLD)as given in the fact sheet of
American Psychiatric Association, (2013) says that the diagnosis of SLD requires persistent
difficulties in reading, writing, arithmetic, or mathematical reasoning skills during formal
years of schooling. Symptoms may include inaccurate or slow and effortful reading, poor
written expression that lacks clarity, difficulties remembering number facts, or inaccurate
mathematical reasoning. Current academic skills must be well below the average range of
scores in culturally and linguistically appropriate tests of reading, writing, or mathematics.

3. Objective of the project

 Prove that digital parameterization of pressure and altitude/tilt patterns in children with
dysgraphia can be used for preliminary diagnosis of this writing disorder. Using our
technique, the previous records of handwriting and current handwriting status can help us
in determining the developmental or sign of dysgraphia.

4. System Design
4.1 Sample Collection
A handwritten test document is taken as input along with the time elapsed to complete
the document. A device is used to take the input as it is being written for online analysis.
4.2 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.
4.3 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.

4.4 Skew Normalization

After line segmentation, different lines may have the different skew angle. The skew
normalization [1] [4] process is based on orthogonal projection length which is shown in Fig
1. To normalize the skew angle, orthogonal projection method that applied on the
segmented lines and. This method efficiently deals with higher as well as smaller skew of
handwriting document.

Fig. 1. Orthogonal projection of image

In Fig. 1, shows the orthogonal projection [1] of the handwriting image. The handwriting is
considered to be written within the rectangle box. The skew angle is Θ and the first and last
black pixels of the word are pointed by Ostart and Olast. The actual length of the word
denoted by Al and projected length is denoted by Ol. The orthogonal projection of the text
line is calculated as
Ol = Olast - Ostart – δ (1)
Here δ is the projection and h denote the actual height of the text line. The relationship
between text line and the projection is
δ = h × cos(900- Θ) (2)
Θ=tan-1((y2-y1)/(x2-x1)) (3)
In this case value of δ belongs 0 to h. If Θ = 0 then value of δ is 0 and if the value of Θ =900
then the value of δ is h.
Here (x1, y1) is the first left down most black pixel of the word that is starting of the
baseline coordinate and (x2, y2) is the last right down most black pixel of the word that is
ending of the baseline coordinate.

4.5 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.

5. Methodology for implementation

• 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, normalization techniques have been applied
to find out space between lines, words and letters in handwriting images.

1. Baseline or Skew Angle

Baseline means the alignment of the text lines and words with respect to the
horizontal direction. Baseline of a writer tells lot of personality with accurate information
about that writer.

2. Writing Pressure
At the time of writing, the amount of pressure put down by the writer on the paper is denoted
as writing pressure. The writer can be classified as light writer, medium writer and heavy
writer based on the writing pressure..
• letter of will, leadership qualities and intelligence
• there are two parts that are ‘t’ bar and stem.
• psychologically the stem symbolizes the energy of individual affirms his
personality, individual assurance and energy flow of the writer
3.The letter ‘t’

•letter of will, leadership qualities and intelligence

•there are two parts that are ‘t’ bar and stem.
•psychologically the stem symbolizes the energy of individual affirms his personality,
individual assurance and energy flow of the writer.

6. Conclusion
Handwriting analysis is an emerging field for personality recognition. The various personality analysis
techniques of this field can be implemented for getting correct personality trait information. Although
graphology is an established science, due to human error and ambiguity in the handwriting sample,
accuracy of handwriting analysis is found to be around 90% correct prediction. Automated personality
identification through handwriting analysis will prove to be a good and helpful system for personality traits
identification. It can find its applications especially in the fields of personnel recruitment, in fields of
marketing, medicine and counselling and also biometrics, forensics studies (analyzing handwriting on
ransom notes in kidnapping cases or blackmailing letters or in the cases of pen poison letters).

References
1. Abhishek Bal and Rajib Saha:An improved Method for Handwritten Document Analysis using
Segementation,Baseline Recognition and Writing Pressure Detection. In:6th International Conference
on Advances In computing & Communications, ICACC 2016,6-8 September 2016, Cochin,India

2. Parmeet Kaur Grewal, Deepak Prashar: Behavior Prediction Through Handwriting

Analysis. IJCST Vol. 3, ISSue 2, AprIl - June 2012. ISSN: 0976-8491 (Online) ISSN : 2229-4333
(Print)

3. Nelson H C Yung, Andrew H S Lai* and Perry ZP Chua: A character segmentation

algorithm for off-line handwritten script recognition. In: Visual communications and image
processing, Taipei, Taiwan, China, 24-26 May 1995, v. 2501, p. 1656-1667

4. C.C. Tappert, C.Y. Suen, T. Wakahara: ON-LINE HANDWRITING RECOGNITION - A

SURVEY. CH2614-6/88/0000/1123$01.00 1988 IEEE

5. R. Plamondon , S.N. Srihari: Online and off-line handwriting recognition: a

comprehensive survey. In: IEEE Transactions on Pattern Analysis and Machine Intelligence (
Volume: 22, Issue: 1, Jan 2000 ) INSPEC Accession Number: 6525227