You are on page 1of 3

www.ingenstech.

com



|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|













Sensorless Human Gesture Recognition System using Image
Processing
Abstract
Hand gesture recognition is a growing and a very vast field of research. It can be used in many places
from biomedical to sports and more. For example if a dumb person needs to communicate
something to someone he can do hand gestures but for the normal person to understand that is
difficult. In such cases the project can be used to convert gestures into voice. Similarly instead of
using keyboard in many computer interfaces and machine control, one can simply use hand gestures
for typing or machine control. In sports the symbols of referees or umpires can automatically be
converted to required format and displayed on the boards. Like such there are lots of applications
for hand gesture recognition.
This project aims at demonstrating real time hand gesture recognition using digital image
processing. This does not involve using a specialized sensors attached to the hand or any special
glove. The gestures can be displayed using ordinary naked human hand. Many hand gestures are
recognized using the project. It produces multiple outputs which can be used depending upon the
application.
Hand gestures are recognized based on many shape parameters such as orientation, centroid,
number of fingers, thumb, etc. It never uses features like color of the skin as it varies from person to
person and also it changes because of illumination. So the algorithm can be used irrespective of
persons and hence it is highly portable.The hand gestures recognition is done using image processing
tools on a MATLB. The result of the recognition can be used for the application. Thus the project can
be highly helpful to dumb people, visually impaired or even for a no frills man machine interface for
industrial automation.
Note: The abstract gives one of the many applications of gesture recognition. It focuses on hand
gesture to voice communication. The same can be implemented for a host of other applications.

Existing System
1. No such system exists in real time
2. Sensors or special gloves are needed
3. Skin color is considered for image processing hence affected by illumination.

Proposed System
1. Uses image processing techniques so no sensors are gloves are necessary.
2. Can be used by any person.
3. Shape based parameters are used and skin color is not considered.

Ingens Tech Private Limited | For Students Use Only | All Rights Reserved
www.ingenstech.com | IEEE Projects 2014 - 2015 abstracts | Final Year Projects

www.ingenstech.com



|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|













Block Diagram


ARM / PIC
A
D
C

U
A
R
T

S
P
I

I
2
C

G
P
I
O

D
A
C

/

P
W
M

Power
& Reset
U
A
R
T

GPIO
ROM
RAM
CLOCK
Voice Output Module
Speaker Amplifier Voice
Processor
S
P
I

LCD Module
S
P
I

LCD
Controller
LCD
Display
USB to UART Interface
USB Port
Converter
IC
UART
Interface
Web Camera
HAND GESTURE RECOGNITION

POWER SUPPLY
Power
Adapter
Reverse
Protection
Voltage
Regulators
Power LED
Ingens Tech Private Limited | For Students Use Only | All Rights Reserved
www.ingenstech.com | IEEE Projects 2014 - 2015 abstracts | Final Year Projects

www.ingenstech.com



|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|

w
w
w
.
i
n
g
e
n
s
t
e
c
h
.
c
o
m

|













Input Data
Hand image captured in real time using a web camera. No pre-stored images used.

Algorithm and Techniques Used
Background separation
Morphological Operations
Contrast enhancement and noise filtering
Blob analysis
Shape Parameter Calculation

Software Tools
LCPXpress IDE or MPLAB IDE
OrCAD For circuit design
Eagle For PCB design
MATLAB for PC programming.
a. Image acquisition Toolbox
b. Image Processing Toolbox

Hardware Required
ARM LPC11U67 or PIC18F45K22
Voice Processor
PC Interfacing
Audio Amplifier
LCD
Web cam







Ingens Tech Private Limited | For Students Use Only | All Rights Reserved
www.ingenstech.com | IEEE Projects 2014 - 2015 abstracts | Final Year Projects

You might also like