Content uploaded by Shailendra Kumar Singh
Author content
All content in this area was uploaded by Shailendra Kumar Singh on Oct 23, 2018
Content may be subject to copyright.
85
Proc. of The Seventh International Conference on Advances in Computing, Electronics and Communication - ACEC 2018
Copyright © Institute of Research Engineers and Doctors, USA. All rights reserved.
ISBN: 978-1-63248-157-3 doi: 10.15224/978-1-63248-157-3-33
Personality Detection using Handwriting
Analysis:Review
Hemlata, Manoj Sachan, Shailendra Kumar Singh
Abstract— Handwriting is one of the unique characteristic to
represent what is in our minds, to communicate with others.
Handwriting shows the true personality including behaviour,
emotional outlay, self-esteem, anger, imagination, honesty, fears,
defences and many other personality traits. It is commonly
known as Graphology. Graphology is the scientific method for
recognizing, assessing and to understand writer’s personality
through the shapes and word patterns in the handwriting.
Personality can be identified through various handwriting
features such as zone, baseline, slant, size, spacing, margin,
pressure etc. Handwriting analysis has wide application in the
various fields such psychology, medical diagnosis, recruitment of
staff, career counseling, writer identification, forensics studies
etc. This paper provides an outline of handwriting analysis,
features of handwriting, its related personality traits, overview of
the handwriting analysis system (HAS) and a literature survey of
the existing papers on handwriting analysis.
Keywords—handwriting, handwriting analysis, personality,
handwriting faetures , graphology
I. Introduction
Handwriting analysis [1]-[3] is a scientific method for
recognizing, assessing and to understand writer's personality
through the shapes and word patterns in the handwriting.
Handwriting shows the true personality of the writer including
behaviour, emotional outlay, self-esteem, anger, imagination,
fears, honesty and many other personality traits. It is
commonly known as Graphology.
Hemlata
Department of Computer Science and Engineering
Sant Longowal Institute of Engineering and Technology, Sangrur
Punjab, India
Manoj Sachan
Department of Computer Science and Engineering
Sant Longowal Institute of Engineering and Technology, Sangrur
Punjab, India
Shailendra Kumar Singh
Department of Computer Science and Engineering
Sant Longowal Institute of Engineering and Technology, Sangrur
Punjab, India
Corresponding Author:
Manoj Sachan
Professional handwriting examiners who identify the
personality through the handwriting samples are called
graphologist. We can interpret the inner psychology and
behaviour through the tone of voice (speech), facial
expression, gestures, posture, and manner of dressing. Often,
the external style mirrors the inward one. Handwriting is also
one of the expressive ways that tells about your nature,
psychology and behaviour of the writer.
Handwriting[1],[2] is unique to each individual. And it will
be same and unique for a writer whether he/she has written
with his/her foot, hand or mouth. Handwriting is written by the
brain, not by the feet or hand. So, the handwriting is also
called as brain writing. Each personality trait has neurological
brain pattern in the human brain. Each neurological brain
pattern design delivers one of a kind neuromuscular movement
which is the same for each individual who has that specific
personality trait. Each stroke or movement in handwriting
reveals a particular personality trait. Graphology is the science
which identifies these strokes in handwriting and describing
the corresponding personality trait. Writer identification is
used for various purposes for example, for security, monetary
activity, forensic & utilizes as access control, analysis of
handwriting documents can be used to judge the culprits in the
criminal justice organizations.
Handwriting analysis or graphology has wide scopes in the
fields such as recruitment, psychology, medical diagnosis,
forensic, human computer interaction. Handwriting represents
the personality and behaviour of the humans so it can be used
in recruitment and staff selection. The handwriting discloses
many things about the writer such as a psychological problem,
morality, hidden talents, health related problems, past
experience etc.[5]. Handwriting analysis can be used to obtain
an insight in to the psyche of the person. Handwriting serves
as one of the diagnostic tool. The handwriting reveals the
psychological and physiological conditions of the patient;
hence it is used as a valuation tool in medical and
psychological diagnosis.
Earlier handwriting analysis was done manually by
spending a lot of time to predict the nature of the person. In
manual analysis, accuracy of the analysis depends on the skills
of the graphologist. The graphologist is also prone to fatigue
when several samples are to be analyzed. For getting a well-
experienced graphologist high cost is incurred. At the other
hand automated handwriting analysis is very fast, accurate,
very low-cost and convenient method in the prediction of
human personality.
This paper mainly focuses on the handwriting analysis
process to detect the personality of a person based on their
handwritten documents. Section II describe the related works
in field of handwriting analysis, section III explains the
86
Proc. of The Seventh International Conference on Advances in Computing, Electronics and Communication - ACEC 2018
Copyright © Institute of Research Engineers and Doctors, USA. All rights reserved.
ISBN: 978-1-63248-157-3 doi: 10.15224/978-1-63248-157-3-33
handwriting features and personality, section IV explains the
process of HAS and section V gives conclusion and future
scope.
II. Related Work
Rahiman A.M et al. [1] proposed an offline, handwriting
analysis tool called “HABIT” which stands for Handwriting
Analysis Based Individualistic Traits Prediction which is used
to detect the personality characteristics of writer from the
given scanned image of a writer‟s handwriting sample.
Champa H N et al. [3] proposed a rule base system whose
input parameters are the pen pressure, baseline, height of the t-
bar, and shape of the loop of letter 'y'. This system predicts the
personality traits from these features. The baseline was found
using polygonalization method, the position of 't' bar and slant
of writing were found by template matching, the pen pressure
using grey level threshold value, the loop of 'y' is analyzed by
Generalized Hough Transform (GHT) technique.
Mukherji S et al. [8] proposed various algorithms to extract
features like baseline, slant, size, margin etc from handwritten
document image for personality detection. Dhadwal A. et al.
[9] proposed a neural network system to know the psyche of
the person through handwriting sample of the writer. This
paper used image processing to detect the pressurized sections,
edge detection, line segmentation and character segmentation
to calculate the slant angles, height and pressure using
thresholding.
Joshi P et al. [10] proposed a machine learning tool with
KNN classifier with incremental learning to improve the
efficiency of the handwriting analysis. The features baseline,
letter slant, height of „t‟ bar, margin are used for personality
analysis. The baseline was calculated using polygonalization
method, margin was calculated using vertical scanning
method, the height of the t-bar on the stem of alphabet„t‟ and
word slant were calculated using template matching.
Mutalib S et al. [11] proposed a system to determine the
level of emotion control of the writer using the fuzzy
inference. The emotion of the writer was analyzed by the
baseline of the writing and Mamdani inference was used. This
system was mainly developed to help the counselor in order to
detect the emotional control of their counselees.
Grewal P. K et al. [12] proposed a system with Artificial
Neural Network to predict the behaviour by using the features
like slant, baseline, pen pressure, letter „f‟, letter „i‟. The
baseline and slant were analysed using the polygonization
method, the pen pressure was analysed using grey-level
threshold value, and letter „i‟ and letter „f‟ were analysed using
template matching.
Cha S.H et al. [13]proposed an artificial neural network to
detect the forged handwriting by proposing a hypothesis that
the authentic handwriting samples has the smooth ink traces
while forged handwriting has wrinkly traces. The feature
which was used includes the wrinkleless, centroid ratio, stroke
width, slant, ascender and decender, the projected histograms
and the gradient histogram.
Coll R. et al. [14] proposed a tool for measuring the active
personality and leadership of the writer. To find the
characteristic trait the writer they used the layout
configuration, shape, letter size, slant and skew angle of lines,
etc. and used a neural network to classify the person qualities.
This paper proposed some new feature like roundness factor
and frequencial analysis of word‟s core region which are
different from one person to another.
Kamath V et al. [15] proposed an automated system to
analyze the behaviour of the person using the features which
includes slant, pressure, size, baseline, margins, number of
breaks, spacing between the words and speed of writing.
III. Handwriting Features and
Personality
There are various features of handwriting which are used
to detect the personality characteristics of the person. These
handwriting features the zone, baseline, pressure, size,
spacing, slant, margin etc. are explained below.
A. Zones
The handwriting is partitioned into three parts called upper
part, middle part, and lower part as shown in Fig.1. The
Personality can be identified by analysing these three zones.
The sections of the three zones are used to define the balance
between the three major areas of ego development of the
writer: the intellectual and spiritual sphere of the individual,
everyday activities and the unconscious intuitive drives [16].
Upper Zone represents: Future, upper body, conscious
spiritual, intellectual and cultural aspirations, mental
perceptions, concepts, fantasy.
Middle Zone represents: Present, middle body,
realistic, practical and social expression of the ego,
emotional expression.
Lower Zone represents: Past, Lower body, memory,
unconscious drives, sensual perception, basic drives,
unconscious drives and biological needs.
B. Baseline
The baseline can be ascending, descending, straight, as
shown in Fig.2. Baseline is used to find the emotional outlay
and nature of the writer. Ascending baseline represents the
person is Optimistic, Hopefulness & cheerfulness, Stay Busy
and Active, Excitability, choleric behaviour. Descending
baseline shows that the person is pessimistic, mental tiredness
of a temporary nature and has digestive trouble. Straight
baseline shows that the person has stable outward behaviour,
straightness, realism, and disciplined.
87
Proc. of The Seventh International Conference on Advances in Computing, Electronics and Communication - ACEC 2018
Copyright © Institute of Research Engineers and Doctors, USA. All rights reserved.
ISBN: 978-1-63248-157-3 doi: 10.15224/978-1-63248-157-3-33
Figure 1. Different zones of characters
Ascending
Baseline
Descending
Baseline
Straight
Baseline
Figure 2. Different types of baseline
C. Slant
The slant of the writing is defined by the direction of letter
slope and is measured by the angle formed between the letter
(down stroke) and the baseline. The slant tells about the
emotions, degree of sentimental control, emotional control and
of the writer. There are three types of slant as shown in Fig.3.
D. Size
Size of the handwriting is measured by the vertical height
of the letters. The size can be large, medium or small as shown
in Fig. 4. Size tells us that how much importance the writer
places upon himself and upon his own actions. It indicates
how the writer will amaze himself upon his environment. For
example, the writer with large size handwriting approaches
life with extroversion and overindulgence and the small writer
with seclusion and shyness.
Vertical slant
Rightward slant
Leftward slant
Figure 3. Different types of slant
Large size
handwriting
Medium size
handwriting
small size
handwriting
Figure 4. Different size of handwriting
E. Pressure
The amount of the force applied at the time of writing is
considered as pressure of pen. It can be heavy, light or
medium as shown in Fig. 5. From the pen pressure we can
analyze the mental energy of the writer. The heavy pressure
depicts the writer is energetic, active, anxious, vigorous,
energetic in everything, angry, alert and punctuate. The
medium pressure shows that the writer's feeling is not very
intense. The light pressure depicts the calmness, passivity,
lack of energy and illness [2].
F. Word Spacing
The space between the ending of first word and starting of
the second word is known as word spacing. It represents the
distance that the writer would like to maintain between
himself and other people as shown in Fig.6. Spacing tells the
closeness of writer with the other people and his intelligence.
Wide spacing shows discrimination, independence, good taste,
exclusiveness, snobbery, pride, has clear thought, ability to
organize his work. Narrow spacing shows Inability to be
alone, poor taste, friendliness, obtrusiveness.
G. Margin
Margin is the amount of space that the writer leaves in the
left side of page or right side of the page or the top or at the
bottom of the page [14]. When writer start writing on the
blank sheet of paper, then they assume some margin space on
that paper sheet and after that start writing. The margin of
various types i.e. top, bottom, left and right margin. Margin
can define the past and future, adjustments, intelligence,
fastness and truthfulness [2].
Heavy pressure
Medium pressure
Light pressure
Figure 5. Different types of pressure during writing
88
Proc. of The Seventh International Conference on Advances in Computing, Electronics and Communication - ACEC 2018
Copyright © Institute of Research Engineers and Doctors, USA. All rights reserved.
ISBN: 978-1-63248-157-3 doi: 10.15224/978-1-63248-157-3-33
Large spacing
Small spacing
Figure 6. Word spacing between words
IV. Handwriting Analysis System
To analyses the personality from handwritten scanned
sample we need to follow some steps as shown in Fig.7.
Figure 7. Personality detection system
A. Scanning
The handwriting samples are taken on a white A4 size
paper. This handwriting sample provided as input to system.
The handwriting sample paper is scanned using a scanner and
stored as image file (JPEG or PNG etc.) format or it can be
taken using a camera.
B. Preprocessing
Pre-processing is done to enhance the image data as input
and also some image features quality is improved for further
processing. Image preprocessing includes noise removal,
binarization, and normalization. Binarization converts the gray
scale image into binary image. Noise removal techniques are
applied to remove the unwanted data and to improve quality of
image. Normalsation is used to remove some of the variations
of handwriting styles and to simplify the shapes of symbols.
C. Segmentation
Segmentation is the process for segmenting the
handwritten page into three different types of segments, i.e.
line, word and character segment. Line segmentation is used to
segment the handwritten image into text lines, lines further
used for word segmentation which are used in feature
extraction process, and character segmentation is used to
segment the words into characters.
D. Feature Extraction
Feature extraction is a process of dimensionality reduction
or extraction of an important data from a high dimensional
input data. The output data is used for analyzing the
personality of writer. The features can be the size, baseline,
slant, margin, zones etc.
E. Classification
Classification is used to recognize the personality traits of
the writers. The various features extracted in the feature
extraction step act as input to the classifier. According to the
feature values the personality of writer is identified. Using
classification methods, classifiers or using rule base system
the classification can be done.
V. Conclusion and Future Scope
HAS identifies the personality of a person. A scanned
image of handwriting sample is given as input and a set of
personality traits are produced as output. The paper studied
has implemented HAS using various handwriting features
such as baseline, size, slant, spacing, margin, pressure etc. and
a more number of features like zone, f, i, speed of handwriting
can be included to make the analysis more accurate. A
Language independent human personality analysis tool can be
created, which can recognize personality from different
handwritings written in different languages. A fully automated
HAS can be formulated that require no human interaction. A
system can be created that recognize the changes in the past
and present personality of a person through the change in the
handwriting. Personality detection using HAS will be a helpful
and efficient system for personality traits classification.
SEGMENTATION
SCANNING
CLASSIFICATION
HANDWRITTEN SAMPLE
PERSONALITY
TRAITS
FEATURE EXTRACTION
baseline
size
slant
zone
PREPROCESSING
binarization
Noise removal
89
Proc. of The Seventh International Conference on Advances in Computing, Electronics and Communication - ACEC 2018
Copyright © Institute of Research Engineers and Doctors, USA. All rights reserved.
ISBN: 978-1-63248-157-3 doi: 10.15224/978-1-63248-157-3-33
References
[1] A. R. M and D. Varghese, “HABIT : Handwritten Analysis Based
Individualistic Traits Prediction,” no. 7, pp. 209–218, 2013.
[2] P. S. Kedar, M. V. Nair, and M. S. Kulkarni, “Personality
Identification through Handwriting Analysis : A Review,” Int. J.
Adv. Res. Comput. Sci. Softw. Eng., vol. 5, no. 1, pp. 548–556,
2015.
[3] H. N. Champa and K. R. AnandaKumar, “Automated human
behavior prediction through handwriting analysis,” Proc. - 1st
Int. Conf. Integr. Intell. Comput. ICIIC 2010, pp. 160–165, 2010.
[4] S. Lee, “Individuality of Handwriting:,” pp. 106–109, 2001.
[5] A. Varshney and S. Puri, “identification on the basis of
Handwriting,” pp. 1–6, 2017.
[6] “The History of Graphology _ British Institute of Graphologists,”
1983. [Online]. Available:
http://www.britishgraphology.org/about-british-institute-of-
graphologists/the-history-of-graphology/. [Accessed: 15-Apr-
2018].
[7] D. J. Antony, “Personality Profile Through Handwriting
Analysis,” pp. 1–118, 2008.
[8] S. Mukherjee, “Feature Extraction from Handwritten Documents
for Personality Analysis.”
[9] A. Dhadwal, S. Alone, and R. Agarwal, “Automatic Emotion
Recognition through,” pp. 811–816, 2015.
[10] P. Joshi, “Handwriting Analysis for Detection of Personality
Traits using Machine Learning Approach Handwriting Analysis
for Detection of Personality Traits using Machine Learning
Approach,” no. November, 2016.
[11] S. Mutalib, R. Ramli, S. A. Rahman, M. Yusoff, and A.
Mohamed, “Towards emotional control recognition through
handwriting using fuzzy inference,” Proc. - Int. Symp. Inf.
Technol. 2008, ITSim, vol. 2, no. 1997, pp. 0–4, 2008.
[12] P. K. Grewal and D. Prashar, “Behavior Prediction Through
Handwriting Analysis Behavior Prediction Through Handwriting
Analysis,” no. June 2012, pp. 13–17, 2016.
[13] K. V, B. Ottappurakkal, S. Suresh, and S. RS, “Personality
Identification Based on Handwriting,” Int. J. Emerg. Technol.
Comput. Appl. Sci. ( IJETCAS ), vol. 12, no. 3, pp. 231–235,
2015.
[14] S. H. Cha and C. C. Tappert, “Automatic detection of
handwriting forgery,” Proc. - Int. Work. Front. Handwrit.
Recognition, IWFHR, pp. 264–267, 2002.
[15] V. Kamath, N. Ramaswamy, P. N. Karanth, V. Desai, and S. M.
Kulkarni, “DEVELOPMENT OF AN AUTOMATED
HANDWRITING ANALYSIS SYSTEM,” vol. 6, no. 9, pp. 135–
140, 2011.
[16] K. Amend and M. S. Ruiz, Handwriting Analysis The Complete
Basic Book. 1980.
About Author (s):
Shailendra Kumar Singh has
obtained BTech degree from UPTU,
Lucknow, India and M.tech in
computer science and engineering
from BIT, Mesra, India. He is
pursuing Ph.D from Sant Longowal
Institute of Engineering and
Technology, Sangrur, Punjab, India
Dr. Manoj Sachan is currently
Associate Professor at Sant
Longowal Institute of Engineering
and Technology (SLIET), India. He
did his B.Tech in Computer Science
from Punjabi University, Patiala,
India. He did M.E in Computer
Science from Thapar Institute of
Engineering & Technology, Patiala
and Ph.D from Punjab Technical
University, Jalandhar, India. His
research interests include
handwriting recognition, natural
language processing and data mining.
Hemlata has obtained BTech degree
from Atal Bihari Vajpayee Govt
Institute of Engineering and
Technology, Pragtinagar, Shimla HP.
She is pursuing M.tech in computer
science and engineering from Sant
Longowal Institute of Engineering
and Technology, Sangrur, Punjab,
India
