International Journal of Recent Technology and Engineering (IJRTE)

ISSN: 2277-3878, Volume-7 Issue-6C, April 2019

Hand Gesture Recognition

Jay Prakash, Uma Kant Gautam

 by these two we save that inputs to databases by assigning

Abstract: As our society and technology is growing day by some task. Now when we give some commands to our system
day, the use of gadgets like mobile phone , computers are also by hand gestures then first machine capture our command as a
increasing. In our daily life we use numbers of way of image then compare this with database and if any image
communication which include speaking, writing, by some body found in database then task assigned to that will be perform.
movement but while in case machine we still stuck with typing or
This idea of hand gesture recognition is very useful for normal
speaking so we need some advancement so that we can
communicate with machine in some body movement also. This user but it is very difficult to implement by programmers but
way of communication in which any type of body movement is still many programmers working on it and we can see many
involved called Gestures. In other words, Gesture is non-vocal gadgets are in market which work on hand gesture. Basically
way of communication which uses hand motion, different the idea is to make a machine intelligent and develop a routine
postures of body, face expressions. So to make machine smart we that the machine understand the human language efficiently.
are enabling our machine to take command by recognizing Gestures could be say easy way of HUMAN COMPUTER
different hand gestures. Hand gestures is used as a input in our
INTERFACES (HCI),so that the computer can understand the
system.
speech, facial expression and human gestures. Gestures are
Index Terms: Computer Vision Based Approach, Hand Gesture non-vocal way of communication like posing a victory sign
Recognition, Human Computer Interface (HCI) , Instrumented from fingers in front of smart phone camera for clicking
Glove , Non-Verbal language.
photos.

II. RELATED WORK

I. INTRODUCTION
Till now there is many hand gesture recognition
In this whole world, there is a vast development of
technologies have evolved. Each technology has its own
computing techniques and due to ubiquitous methods of
advantages and disadvantages. First technique uses wires in
computing, current user interaction with the pointing and
which for taking input users tied up with wires to connect with
positioning devices such as mouse, keyboard and pen are not
computer system. Disadvantage of this technology is user
that sufficient. These devices are only limited so the
cannot move freely and it was difficult to work with wires
commands set is also limited. Making use of body parts for
without getting any harm. Instrumented gloves are used in this
interaction such as use of hands, is a better option. Hands can
technology for hand gestures. In instrumented glove some
be used as an input device for providing natural interaction (1)
sensors are used which gave reading to the system when the
To detect hand gesture recognition (2) there are
hand or fingers do any motion. Data glove optimal results but
basically two ways. First one is using instrumental and other
they are very expensive that's why normal person cannot own
one is vision-based recognition. In instrumental glove, there is
it and also it can easily disturbed so managing it also difficult.
a glove (3) in which some sensors are fixed on the glove and
As wire are very difficult to handle so these wires are replaced
to take input from that glove, some person must wear it and
with optical markers. In this also user wear glove but working
then according to hand movement or finger movement the
in glove is now different now an infra-red light source placed
input for hand gesture are produced. This instrumented glove
at one end and a screen at other end. when hand or finger
work on photocell. As the light hits on photocell cause of
moved then intensity and direction of light changes and that's
movement of hands and fingers and by this input also vary.
how for different gestures different light intensity and
Instrumented glove is less used as its cost is high and while
according to that gestures are recognized. This optical way is
taking inputs in this technique more chances of error as to
also good but it's configuration is very complex. After this
hold hand in same for long time is difficult. To overcome this
Image based hand gesture recognition technique introduced in
approach vision based approach came into existence.
which image is processed like image shape, size , color etc.
In vision based (4), inputs are taken from a webcam
these features are used to recognize the gestures. But working
and but this method is more complex than glove method
with is bit difficult as compare to previously proposed one
because in this number of camera and their position all
because in this to detect color, shape, size is bit difficult as
matters. In this approach , visibility is main component.
background and skin color could be same and different person
Capturing image from stream then detecting skin and
could have different size of hand. Illumination also have
differentiate it from background as background could also
important role in image based because of illumination some
have same color as skin so all these things make this vision
noise could be removed by image while may be on less light
based complex. But it is still quite popular because chances of
the noise could be added to image which change the result.
error less and efficiency is high. After taking input from either

Revised Manuscript Received on December 22, 2018.

Jay Prakash, Department of Computer Science and Engineering, Madan
Mohan Malaviya University of Technology, Gorakhpur.
Uma Kant Gautam , Department of Computer Science and Engineering,
Madan Mohan Malaviya University of Technology, Gorakhpur

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: F90300476C19\19©BEIESP 54 & Sciences Publication
 Hand Gesture Recognition

After all these techniques the Vision Based approach A. Research Status
introduced which worked mainly on shape. As every person Many researchers and students work on Hand Gesture
have one thumb and four fingers in normal condition so this Recognition so they use different approaches to make their
technique is more successful for hand gesture recognition system efficient but in each work there is always some
than other techniques. This method is very successful also limitation which cannot be nullify so to make our system
described in this paper Mehra et. al. (5) for hand gesture efficient we are taking help of other research papers and
recognition. analyze where the author made errors. Like some researchers
According to Anant Atray (6) in his paper "Automatic used algorithm based on Genetic Algorithms (8) and also
Hand Gesture Recognition" he is using best algorithm for algorithms based on neural network (12) these two methods
hand gesture recognition which give optimal result and are good in gesture observation purpose but inefficient as
complexity is very low but there is one limitation in his compare to Hidden Markov Model. Both of take huge amount
project that if the skin color is same as background or of time to train the system and also large amount of training
environment where hand present then his system is not able to data that's why we are using algorithm based on Hidden
distinguish between skin color and background which is Markov Model as it comparatively faster than both GA and
major issue cause it can change our entire result and can lead Recurrent Neural Network.
to false output. So in my project we are using YUV color There is also some problem with many system which
space and camshift algorithms which help to differentiate are using histograms and also skin detection problem (13).
between skin color and background. Skin problem issue is major because in this noise increases
According to T. Freeman et. al. (6) they are using the drastically if skin color and background color is similar that's
orientation histogram to classify different gestures and
why we are using YUV color space and Camshift algorithm
interpolation. Histogram is very simple and powerful as even
which to distinguish between background and skin color.
in different lights and colors the gesture recognition will be
always correct. The main advantage of histogram is in
dynamic gestures because in dynamic gesture the movement III. HAND GESTURE RECOGNITION BASICS
of hand is changed over time period so according to this paper
histogram helps to find dynamic gestures very easily. But only A. Human Computer Interface(HCI)
limitation in using histograms that similar gestures can have Developing some technology that let human to interact or
different types of histogram orientations and similar communicate with machine is Human Computer Interface. To
histogram orientation can represent completely different interact with computer human can use many things like hand
gestures so this is bit confusing in gestures using orientation gestures , eye recognition or some other device like
histogram (6). instrumented glove etc.
As gesture recognition is very useful for nowadays so
some researchers also forming Gesture recognition using B. Gestures
recurrent neural networks. According to Murakami et.al. (7) Gesture (14) is also a way of communication like we
recurrent neural network also has its own advantage in gesture communicate in daily life. But in daily life, we used verbal
recognition as in sign language to differentiate between words and vocal communication while gesture is non verbal way of
like trained-untrained , mom-dad , present-absent are very communication or we can say it is a silent communication.
hard. So in this neural network help a lot cause using this our Gestures constitute every type of communication except
system can capture past history to identify the word are used verbal and written. The other way of communication apart
in pairs that are confusing. But recurrent neural network in from that is making hand movement , giving facial
gesture recognition has its own disadvantages it is very slow expressions , giving different type of body postures these all
means even to learn only 10 to 20 words while training the are gestures.
system , the system can take 4 to 5 days which is large amount We can say gesture is the body language that means to
of time that's why in my system we are not using neural communicate with others by just making the movement and
network for gesture recognition (7). motion of body parts. Gesture is very quick communication if
Some researchers used algorithms based on Hidden both parties understands the gesture meaning same. By this
Markov Model and some use the algorithms based on Genetic we can also show our emotions by just some movements.
Algorithms. According to X. H. Wang et.al.(9) Genetic According to movement gestures are of two types :
Algorithm(GA) convert capture input as a discrete points and a. Dynamic Gesture - Change over period of time (16).
after that convert the problem of recognition gesture image Examples of dynamic gesture :
into problem of combinatorial optimization that include Waving of hand means “goodbye”.
discrete points and then gesture recognition algorithm applied
to detect gestures. But This genetic algorithm have
disadvantage that it require high amount of training data and
long period of time. So, we are using Hidden Markov
Model(HMM) is our work because learning algorithms used
for HMM is very efficient. Using HMM learning can take
place directly from sequence of raw data. Our work using
HMM which also advantage over genetic algorithms that it
uses less number of training data as compare to GA and also
training time is less in HMM.

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: F90300476C19\19©BEIESP 55 & Sciences Publication
 International Journal of Recent Technology and Engineering (IJRTE)
ISSN: 2277-3878, Volume-7 Issue-6C, April 2019

b. Static Gesture -- observed over a spurt of time (17). environment. The virtual environment that are using in
It is type of static gestures (12) which means "stop". present days can be displayed on screens and allow user to
implement all of system applications through it. We can
divide Virtual reality into:
a. Forming a real environment copy or simulation for testing
and training of software , project or systems. We can also
use this for education purpose.
C. Gesture Recognition b. We can develop a environment which is like real life places
but actually doesn't exist . For example in Games like
Understanding a full message, interpretation of all static
Pubg and GTA Vice city we see many maps or places
and dynamic gestures is necessary for over a period of time.
which is same as real life places but in actual don't exist.
This complex process is GESTURE RECOGNITION.
example of gesture: Shoulder Shrugging D. Sign Language
This language is different from our daily language we use
verbally or which we speak. This sign language (2) includes
all the non- verbal way of communication (5) likewise the
movement of body on any statement made, gesture made by
hands to illustrate things to some person or making
expressions by face, All these things are included in sign
language. This language is frequently used by deaf and mute
persons.
IV. PROPOSED WORK
There are many use of Hand Gesture Recognition and that's
the reason significant number of Gesture Based application
evolved, Some of them are listed below :
A. 3 Dimensional Geometry Design
Auto CAD (computer aided design) is an Human Computer
Interface which is used for designing and drafting of 2
dimensional and 3 dimensional images. By using mouse and
keyboard it is difficult for an programmer or user to design 3D
design because making a 3D images involves all 6 Degree of
Freedom(DOF) and allocating points in space using mouse is
very hectic and complex. Now CAD provides facility to
translate points or rotate points of image in any direction.
Using this we can also see image from each and every
direction according to our requirement to analyze it (9).
Massachusetts institute of technology (10) has come up with
the 3D RAW technology that uses a pen embedded in
polhemus device to track the pen position and orientation in
3D.
B. Telepresence
Tele Presence (2) is way of communicating to person
which are not in same room by using internet services like
digital video or 3D visualizer. Tele Presence advantage is that
any user can remotely present on location where he is permit Fig 1. Gestures used for English Alphabet
to connect itself. For instances, in many companies, if
employees of company in different countries or in different
Before this language it is very difficult to communicate
places then meeting still can be done by using video camera,
with some physically challenged persons. Some time they
microphone and video screens. The Tele Presence include
understand you but cannot express themselves and other time
collaborations especially instructions often, depends on the
they can express themselves but cannot understand you.
physical act of one person showing another person how to do
After this language, by watching, a deaf person can
something and even if your Tele presence robot has an arm of
easily understand what you want to say and a mute person can
two it may not be at all intuitive for a remote user have
even express themselves.
effective direct interactions.
C. Virtual Reality
Virtual reality (11) is computer generated 3 dimensional
environment with the help of software for users so that user
can run its program, test its system by assuming it be a real

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: F90300476C19\19©BEIESP 56 & Sciences Publication
 Hand Gesture Recognition

Even in some news channels we can see one side reporter E. Linear Fingertip Model
speak and according to that reporter report a person besides it The linear finger model are finger movement are assume
use sign language to sparse that report in deaf person too and the linear rotational movement. Finger tissue modeling
also by that a normal person can learn how to correctly use requires linear deformation models The Fingertip model
that sign language. posses capability of linear movement of finger pad, suitable
fingertip model which can respond for every type of hand and
V. ALGORITHMIC TECHNIQUES USED FOR fingers because person to person size of hand, size of fingers ,
RECOGNITION OF HAND GESTURES their joint angle and their movement could vary. Linear
To collect raw data we use either vision or glove based data Fingertip model enforced to various type of deformation
collection system and various algorithms (14) used in order to effects which are taken through the cameras during computer
collect the raw data smoothly and correctly. vision based tracking, like the video stream is captured with
Various algorithm used are : less control by which boundaries condition are hard to extract.
Gesture Recognition Steps
A. Template Matching
Step 1: First image is captured taken from stream using
The template matching method for hand recognition camera which is working by OpenCV.
postures and gestures recognition used experimental method Step 2: Now the image is processed that what data it is
to know the required number of template of a certain gesture containing.
to be taken that should be saved on the database for the Step 3: On data algorithms are applied to find optimal
matching process of the algorithm. If the system will not be gesture.
able to detect and recognizes the gesture given with the Step 4: Obtained gesture is now compared with gesture
templates an additional templates must be trained and stored dataset.
in the database until the system accurately recognize the Step 5: When if any dataset gesture is matched with input
gestures. The proponents will sum up all the time in second gesture then the result will displayed on screen or gesture is
under a certain number the same number of template gesture. recognized.

B. Analyzing Extracted Features

When we recognizing some image or some pattern then VI. HAND DETECTION AND RECOGNITION
first we gather information on low level or we can say we MODELS
identify features of raw data and according to features we
categorize the inputs is called Feature Extraction. A. Hidden Markov Model
A tough feature will be invariant, meaning that if the image In Hidden Markov Model(HMM) (14), gestures are
is rotates shrunk or enlarged, or translated, the value for the
captures from every picture that makes a video and in this skin
recognize hand gestures and postures.
color blobs are tracked corresponding to hand-face likely on
C. Active Shapes Model the face of user. Hidden Markov Model is a Markov Model
The Active Shape Model is training is done by manually which is made by using unobserved states. These unobserved
drawing 3 dimensional surfaces images. The Active shape states is also called Hidden states. In hidden Markov model
Model (9) find the difference in the data by which it is training ,the output is visible which dependent on the states while state
using Principal Component Analysis . is not visible directly.
Active shape model find best match for object or input
B. YUV color space and camshift algorithm
outline according to data stored in database during training
process. It iteratively checks every image and frame then try How hand gestures are going to be recognized is dealt with
to find the best image which have similar control points as by this algorithms. Following steps are included in order to
training data image. recognition of hand gesture :
1) System Camera or digital camera first takes input of
D. Analyzing Principal Component frames which are making video stream of hand motion
Analyzing principal component is a important technique gestures.
to understand in the field of statistics and data science 2) The existing fames into video stream that are going to be
interrelated variables is called Principal Components our input are grasped and then process of segmentation is
Analysis. When putting the students online to technical didn’t performed and this is based on YUV color space.
fully address our needs and provided conflicting reducing the 3) The YUV color space system is used for distinguish
dimensions of the features spaces is called dimensionality intensity and chrominance of frames that are captured from
reduction information.PCA as accessible as possible the video frames. In word YUV, Y specifies the intensity in video
algorithm well cover is pretty technical. Some all of the frame and UV indicates chrominance components in video
following will make this article and PCA as a method easier to frames.
understand: matrix operation/ linear algebra (matrix 4) After all this CAMSHIFT algorithm is used for hand to
multiplication, matrix transposition, matrix inverse, matrix bifurcate it from body as hand and other body parts colors are
inverses, matrix decomposition, eigenvectors / eigen values) . same.
When the dealing image the image have position
sensitivity , the joint angle between hand and fingers their
orientation all these come in Principal component analysis.

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: F90300476C19\19©BEIESP 57 & Sciences Publication
 International Journal of Recent Technology and Engineering (IJRTE)
ISSN: 2277-3878, Volume-7 Issue-6C, April 2019

To segment the hand from other body parts we use logic that finger in one hand and our hand's thumb which have joint with
is the hand is largest connected region. our index finger is also detects by this sensor based glove
5) Hand position is calculated in each frame of video stream
and to calculate it centroid of hand is calculated . To calculate
centroid ,from initial to last position is calculated.
6) To find the path of our input i.e. hand movement we have
to join the all centroid points which form a trajectory and by
using all these procedures we can track the hand movement.

C. Naive Bayes Classifier

Naive Bayes Classifier basically works on Bayes theorem
which is collection of different classification algorithms. This
is collection of different algorithms which is based on same
logic that is every feature should be classified and these
features should not be depend on each other.
The Bayes Theorem (18) used find the probability of an Fig. 2 Instrumented Glove
event if any other given event already has been occurred.
P(E|F)=P(F|E)P(E) / P(F)
B. Computer Vision Based Approach
This method is used for recognition of static hand gestures.
It is very efficient and quick method for recognition. As Naive While using Instrumented gloved there are many
Bayes collect different input according to their features so difficulties faced by user like user have to always wear that
glove while collecting the hand gestures. It is very difficult
here different gestures are classified according to their
to give same posture same again and again because angle of
different geometry and orientation. Every frame in the video
hands and finger could be vary each time which make input
sequence is used to extract gestures in which background is
gestures two different gestures.
static.
On time of collecting the data the user is not allowed
VII. COLLECTING DATA FOR HAND GESTURES to make movement for some amount of time until input is
not recorded correctly. So, Computer Vision Based (19)
We can collect input for this by mainly Three type. came into existence. To collect data by using Vision Based
First Way is that user can wear the device which take Approach for Hand gesture and movement there are four
input. This device detects every angle making by hand and all constituent which make system as whole:
the degree of freedom. This input device is type of glove
which detects the position of hand in space and according to Number of camera used and where to place them is first
that raw data produced. challenge in this approach. As we have to place cameras
The Computer Vision Based technology is another such that each and every movement of hand could be
way of collecting data in this camera takes images of hand register. Main problem that encounter in this type of
gesture which are given by during input. These stream of approach that hand movement is not visible by cameras and
images is sent to routine which is for image processing by they also fail to record hand movement so this is primary
which what input given by user by their gesture could be concern in this method.
identify.
The hybrid way is last approach is used to take input
as in this Vision and Glove Based both are used Image Capture Image Feature
simultaneously which make raw data more accurate. Main By camera Processing Extraction
reason using this hybrid way is when there is some error by
one method the other will compensate by its correct data.
Result Noise Feature
A. Instrumented Gloves
filtering Matching
While designing the these glove (3) there are some circuits
and some sensors are fixed on glove which detects every
single motion made by hand or by even fingers. Fig. 3 Flow Diagram of Working of Computer Based
These instrumented gloves design and sensors used in it Approach
can be differ according to use or by companies made it. That's The next (1)thing that constitute in this technology is
why it is not necessary every glove work same as other. extracting the simplest hand gesture data that have accurate
In this glove there are some sensor whose working is Light visibility in camera.
based. In light based a tube is used which is flexible so that it
can cooperate with hand movement in placed at one side and
on the other side photocell is used. As we move or make
movement by finger the light that strike on photocell changed,
by this finger flexion could be measured. Meta
carpophalangeal of finger which is basically joints of all our

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: F90300476C19\19©BEIESP 58 & Sciences Publication
 Hand Gesture Recognition

Next constituent in this approach to captures every image System performance enhance when we train our system
from stream of images which are running. The last with maximum number of datasets. As we want to be our
constituent to work this device efficiently some efficient system more and more reliable for that we should train it with
algorithm should be applied in it which can take input, maximum number of datasets.
compare it with entries in database and then give result in
minimum time. Here we are showing how a Vision Based REFERENCES
Approach work(20).
1. Matthias Rehm, Nikolaus Bee, Elisabeth André. Wave Like an
C. Comparison Between both Models(2) Egyptian – Accelerometer Based Gesture Recognition for Culture Specific
Interactions. s.l. : British Computer Society, 2007.
https://ewic.bcs.org/content/ConWebDoc/21329
Glove 2. Sanjay Meena. A Study on Hand Gesture Recognition Technique(Thesis
Vision-Based
Methods Based Submitted in partial fulfillment of the requirements for the degree of Master
Model of Technology in Telematics and Signal Processing)
Model
3. Thomas G. Zimmerman, Jaron Lanier, Chuck Blanchard, Steve Bryson
Price High Less and Young Harvill. A HAND GESTURE INTERFACE DEVICE.
Alleviation Less High .(published in CHI '87 Proceedings of the SIGCHI/GI Conference on Human
Factors in Computing Systems and Graphics Interface
for User Pages 189-19 https://dl.acm.org/citation.cfm?id=275628
Hand Shape To be Can differ 4. Toward a vision-based hand gesture interface. Quek, F. Singapore : s.n.,
August 23–26, 1994. Proceedings of the Virtual Reality System Technology
same Conference. pp. 17-29.
Calculation Complex Less complex 5. Hand Gesture Recognition for Human Computer Interaction. Mehra,
than GB Meenakshi Panwar and Pawan Singh. Waknaghat, India : s.n., November
2011. IEEE International Conference on Image Information Processing
Efficient Lesser High (ICIIP 2011).
6. AnantAtray. Automated Hand Gesture recognition.
7. William T. Freeman, Michal Roth,. Orientation Histograms for Hand
Gesture Recognition. Cambridge : s.n.
8. Kouichi Murakami, Hitomi Taguchi. Gesture Recognition using
VIII. CHALLENGES IN HAND GESTURE Recurrent Neural Networks. Japan : s.n.
RECOGNITION 9. X. H. Wang, H. J. Bao,. Recognition Based on Adaptive Genetic
Algorithm. Journal of Computer-aided Design & Computer Graphics. 2007.
Hand gesture recognition confronts many challenges (20), 10. A framework for hand gesture recognition and spotting using
these challenges are: sub-gesture modeling. M. R. Malgireddy, J. J. Corso, S. Setlur, V.
Govindaraju, and D. Mandalapu. 20th International Conference on Pattern
A. Changed illumination: Change in light effects can affects Recognition (ICPR '10). p. 3780.
1our gesture input as it can change extracted skin region. 11. R. Pradipa, Ms S. Kavitha. Hand Gesture Recognition- Analysis of
B. Rotation Problem: Problem of Degree of freedom. if Various Techniques and their algorithms. 2006.
12. Dynamic hand pose recognition using depth data. P. Suryanarayan, A.
degree of freedom is changes then our gesture input may Subramanian, and D. Mandalapu. August 2010. 20th International
differ and by this output could vary Conference on Pattern Recognition (ICPR '10). pp. 3105–3108.
C. Distinguishing Problem: If with hand there is other things 13. Real-time hand status recognition from RGB-D imagery. A. D.
Bagdanov, A. Del Bimbo, L. Seidenari, and L. Usai. November 2012. 21st
which have shape and color like our skin but that is not International Conference on Pattern Recognition (ICPR '12). pp.
our gesture input then it can create problem for system to 2456–2459.
distinguish between our hand gesture input and 14. Vladimir I. Pavlovic, Rajeev Sharma, Thomas S. Huang. Visual
Interpretation of Hand Gestures for Human-Computer Interaction;, . s.l. :
background.
IEEE Transactions on Pattern Analysis and Machine Intelligence,
D. Size Problem: This As per human being , we have different 1997.https://pdfs.semanticscholar.org/8dd3/d5ef2e1c40433417cd90e21ce9
shapes and sizes of hand like small child have small a0468a7626.pdf
hands and adult have big hands so it can create problem 15. Simon, Haykin. Neural Networks: A Comprehensive Foundation. s.l. :
2nd edition Prentice Hall PTR.
for system. 16. Virtual Reality. agentic.ca. [Online]
E. Position Problem: While giving input if hands position http://www.agentic.ca/virtual-reality.
differs like hand placed in corner of screen or all dots 17. Stuart, A. and Ord, K. Kendall's Advanced Theory of Statistics:
Volume I—Distribution Theory. s.l. : Edward Arnold, 1994.
which detecting hand position doesn't lies on hand then it 18. Xu, Yanan and Dai, Yunhai. Review of Hand Gesture Recognition
can create problem to capture input from user. Study and Application. Contemporary Engineering Sciences. 2017, Vol. 10.
Available: .https://doi.org/10.12988/ces.2017.7326
19. Model-based hand tracking using a hierarchical Bayesian filter.
IX. CONCLUSION Stenger B, Thayananthan A, Torr PH, Cipolla R. Sep 2006. IEEE
Hand Gesture Recognition is an important Human Transactions on Pattern Analysis and Machine Intelligence. pp. 1372-84.
20. Shigeo, Abe. Support Vector Machines for Pattern Classification.
Computer interface for interaction between living thing like Kobe University.
human and machine system. Hand Gesture Recognition
System works like this : first user give input to the system by
making hand gestures, then system scanned the gestures by
using cam or sensor and deducts it into signal and passes the
program, now its program responsibility to first accept the
signal then examine what is the input given using gestures,
then check if there is any corresponding data is saved into
dataset then we will get our result.

Published By:
Blue Eyes Intelligence Engineering
Retrieval Number: F90300476C19\19©BEIESP 59 & Sciences Publication