Personality Prediction using HandWriting Analysis

Submitted by:-
Aanchal Khandelwal (10103600)
Manvi Garg (10103622)

INTRODUCTION
 Your brain guides your hand. Everything put on paper
is a result of a two-way circuit between your brain and
the motor reflex muscles of your hand.
 In our Project we are analyzing the handwriting of
individual to depict the personality of that person.
Handwriting becomes a Polygraph or Oscilloscope
read-out of your "complete self."
To you, it's just handwriting, but to a handwriting
analyst, it paints a picture of the person "behind the
pen."

Handwriting Analysis- What is it?
 I know it seems impossible, but a trained graphologist and by
using today’s technology it is possible we can gather an
astonishing amount of information about the writer just from
analyzing their handwriting.
 Besides creating a complete personality profile, many other things
are revealed in your handwriting, such as health issues, morality,
past experiences, hidden talents, mental problems-- to name just
a few.
Handwriting analysis, or
graphology, is the science
involved in producing a
personality profile of the writer by
examining the characteristics,
traits and strokes of an
individual's handwriting.

PROJECT
 Prediction of Personality by handwriting analysis. Every
individual have different style of writing as his/her brain have
different ways to show what’s going on in his mind so we
came across with a idea to make this project.
 One can easily give the digital sample of his/her handwriting
to a computer and it calculates the features using the image
processing techniques and predicts the nature of the writer.
 To make this computerized we considered Different types
of features and studied them thoroughly
We used python as our platform.

 Allograph and allograph combination analysis is a scientific
method of writer identification and evaluating the behaviour.
 To make this computerized we considered six main different types
of features:
(I ) size of letters (II ) slant of letters and words (III ) baseline
(IV) pen pressure (V) spacing between letters ( VI ) spacing
between words in a document to identify the personality of the
writer.
 Segmentation is used to calculate the features from digital
handwriting in which the letters or words are segmented using
image processing and applying various steps.
Handwriting image is uploaded by
the user or individual and then the
image is processed .

Research analysis
S.NO. WRITING CATEGORIES PERSONALITY BEHAVIOUR
1 LARGE LETTERS Likes being noticed, stands out in crowd
2 SMALL LETTERS Introspective, not seeking attention, modest
3 MEDIUM LETTERS Adaptable, fits into a crowd, practical, balanced
4 RIGHT SLANT Sociable, responsive, interested in others, friendly
5 LEFT SLANT Reserved , observant , self-reliant
6 VERTICAL Practical , independent, controlled ,
7 LIGHT PEN PRESSURE Strong emotion , successful, emotion last for long time
8 HEAVY PEN PRESSURE Try to avoid energy draining situations
9 FAR SPACING LETTERS Openness of sentiment and intelligence
10 CLOSE SPACING LETTER Closeness of sentiment and intelligence
11 FAR SPACING WORDS Give me breathing space
12 CLOSE SPACING WORDS Wish to be with others , intrusive
13 RAISING BASELINES Optimistic, upbeat, positive attitude
14 FALLING BASELINES Tired , overwhelmed , pessimistic , not hopeful
Psychological personality traits
predicted by various handwriting styles
of writer.

 1. A writer with heavy pressure and a vertical slant.
Heavy pressure= strong emotions and vertical slant= trying to hold
emotions back. This writer is usually the one who keeps his wits about
him. When "all hell breaks loose," his head rules. He will not be as
emotionally responsive as a right slant writer and will "keep cool". This
"heavy pressure-vertical slant" writer won't be able to keep his cool all of
the time. When his emotions get the best of him he may "fly off the
handle" at unpredictable times.
 2. A person with light pressure writing (not much emotional or physical
energy) and a left slant (tries to avoid emotional situations). This person
will be emotionally withdrawn, cold, indifferent and self centred .
 There are many steps involved in creating a complete personality profile.
Using these first two steps you can begin to put together your own
"profile" which can be useful in both personal and professional
relationships.
Now let's mix some of these traits and see what we
come up with.

How we do it ?
We follow certain steps to reach our goal
 First an image is uploaded by the user it should be on
plain paper and should be in writers handwriting only.
 To get the good result we need a good image or by using
threshold we remove the noise from the image now as the
image is free from noise it can be looked upon further .
 Now image is pre-processed and using image processing
techniques we will segment the images in different parts
first we will segment the image in word and then we will
segment each character of image this will give extra area
to look on the nature of the person personality

Now as the image is segmented and we have every
character in the image we would now extract the
different outputs from the image we would check
each feature. We would check slant in the letters
i.e they are slanted right or left or vertical. We
would check the baseline angle of the lines in the
image are these lines straight that means whether
person right in straight line or not. Now after
checking other pre specified features we would
come with the result of the person personality.
Image processing play a crucial role in the project
as it help us to get the segmented part of the
image. These segmented parts of the image are the
basic of what we actually need.

 Segmentation is used to calculate the features from
digital handwriting, contour selection and canny is
used to select the outlines of words and edges of
lines. Finally, all extracted features are fed to svm and
ann which outputs the behaviour of the writer. The
proposed method tells us that which method is
efficient and effective with the actual real graphology
method.

SVM
 support vector machines are supervised
learning models with associated
learning algorithms that analyze data and recognize
patterns, used for classification and regression
analysis.
 We used RBF to give input to the SVM basically SVM
take input as a function .
 SVM give the better result and Exact result as
compared to the ANN

RBF
 A radial basis function (RBF) is a real-valued
function whose value depends only on the distance
from the origin
 Sums of radial basis functions are typically used
to approximate given functions . This
approximation process can also be interpreted as a
simple kind of neural network. RBFs are also used
as a kernel in support vector classification.

Books for Handwriting Analysis

It revels
 Your Personality
 Your emotions
 Your likes and dislikes
 Your Strengths and weakness
 Your intellect and much more.

Tools , Platform & Software needed
 For our project we need python 2.7 and above
 Operating system may be windows or Ubuntu or any
other Operating system supporting Python
 Other library used in the project are
Numpy - Number Python , a powerful N dimensional
array object.
Open CV- library for image processing
Mat lib- Mat plot library.
My sql db- used to connect to database.

Novelty/benefits
 To make this computerized this we considered few old and
some new features to handwriting analysis:
 size of letters, baseline, slant angle of letters to the
baseline, word spacing
 Dots of I’s, loops of y ’s and f ’s , height of T-bars in a
document to identify the personality of the writer. For
extracting the features we used various functionality and
processes.

Benefits:
 With increase no. of features, we will get better results
and effective.
 Also, on comparing two classifier output, we’ll see that
which classifier is better in terms of efficiency and
matched to real world scenarios.
 We’ll check the compatibility between two people, so it
going to tell they are good for team work or not.
 We can check that if a person is able to handle pressure
or not in case of difficult situation.

Future Scope
It would be proved to be good application for those psychologists and
doctors who want to treat their patient and predict their basic nature and
how it can be improved.
It can be used for employees, so we can see their particulars interests so
we‘ll get benefits from their talents.
We’ll calculate the average width and height of letters of a person’s
handwriting, and compare the result with other persons.
We can send handwriting analysis of the person to their phone so that
person can get his information any time.

References
 1. Shitala Prasad, Vivek Kumar Singh, Akshay Sapre “Handwriting Analysis based on Segmentation Method for
Prediction of Human Personality using Support Vector Machine”, International Journal of Computer Applications,
Volume 8– No.12, October 2010.
 2.Abdul Rahiman M,Diana Varghese,Manoj Kumar G “HABIT:Handwritten Analysis Based Individualistic Traits
Prediction”, International Journal of Image Processings, Volume 7, October 2013.
 3.Perut Boribalburephan and Boonnatee Sakboonyarat , “An Algorithm Development for Handwritten Character
Recognition by Personal Handwriting Identity Analysis [PHIA]”, Knowledge and Smart Technology(KST),2012 4TH
International Conference, Pages 6-10.
 4.Abedelkadir Asi,Raid Saabni and Jihad El-Sana, “Text Line Segmentation for Gray Scale Historical Document
Images”Workshop on Hstorical Document Iaging and processing pages 120-126,2011.
 5.Pankaj kumawat,asha khatri,baluram nagaria, “Comparative analysis of offline handwriting recognition using
invariant moments with HMM and combined SVM-HMM classifier”, Communication systems and network
technologies(CSNT),2013 international conference pages 140-143,2013.
 6.Vassilis papavassiliou,themos stafylakis,george carayannis, “Handwritten document image segmentation into
text lines and words.”, Volume 43,issue 1,pages 369-377,2009.
 7.Champa H N,K R AnandaKumar, “Automated Human Behavior Prediction through Handwriting Analysis” ,
Procedings in integrated intelligent computing(ICIIC),2010 FIRST INTERNATIONAL CONFERENCE,PAGES 160-165,
2010.
 8.Tal Steinherz, David Doermann, Senior Member, IEEE, Ehud Rivlin, Member, IEEE, and Nathan Intrator, “Offline
Loop Investigation for Handwriting Analysis”, IEEE Transactions on pattern analysis and machine intelligence, vol 31,
no.2, 2009
 [9] Champa H N, Dr. K R AnandaKumar, “Artificial Neural Network for Human Behavior Prediction through
Handwriting Analysis” , International Journal of Computer Applications (0975 – 8887) Volume 2 – No.2, May 2010.
 [10] Tristan Wright, “Handwriting Recognition with Artificial Neural Networks and Open CV”.
 [11] D. John Antony, O.F.M.Capp, “Personality Profile Through Handwriting Analysis” , A Textbook of Handwriting
Analysis : Anugraha Publications , Anugraha (Tamil Nadu Capuchin Institute for Counselling, Psychotherapy and
Research), September 2008


Personality Prediction using HandWriting Analysis

More Related Content

What's hot (20)

Similar to Personality Prediction using HandWriting Analysis (20)

Recently uploaded (20)

Personality Prediction using HandWriting Analysis