ISSN: 2278 – 909X

International Journal of Advanced Research in Electronics and Communication Engineering (IJARECE)

Volume 5, Issue 7, July 2016

Smart vision for the blind people

R.Mohanapriya,U.Nirmala,C.Pearlin Priscilla

II. LITERATURE SURVEY

Abstract-The spread of visual impairment is a very
sensitive issue worldwide. Blind people have to play in The paper as in [2] by G. Balakrishnan et al, (2007)
the everyday actions of different difficulties. These describes about theVisually impaired to find their
include the difficulties of moving in complete
autonomy and the ability to seek and recognise navigation as they often lack the needed
objects. Until a decade ago, the only aid that a blind
person has used are sticks, guide dogs accompanying information for by passing obstacles and hazards.
persons or to move. In the last decade, electronic Electronic Travel Aids (ETAs) are devices that use
devices have been introduced into the world of the
blind in order to facilitate the lives of these people. sensor technology to assist and improve the blind

Index terms- feature matching, object detection, user’s mobility in terms of safety and speed.
surfing, template, traffic recognition.
The paper as in [6] (Peter B.L.Meijer et al, 1992)
I.INTRODUCTION
says about the arrival of fast and cheap digital
Various techniques are there which a visionless electronics and sensory devices opens new
person commonly uses such as white cane or pathways to the development of sophisticated
walking cane for navigation. But still the blind equipment to overcome limitations of the human
people cannot cross the road on their own and senses. This paper addresses the technical
depend on others to cross the road. So, on taking feasibility of replacing human vision by human
this into account we decided to design a product hearing through equipment that translates images
which would help the blind people specifically into sounds, which could one day become relevant
cross the road as in [3]. This project involves for the visually impaired.
helping the blind to recognise traffic signal pattern
The paper as in [7] by F.Hongand, A.Chekima,
as well as obstacles around and to cross the road
(2001) says about the goal of Blind Aid project to
without depending on others.
develop navigational assistance technology for the
The project also aims at implementing GPS blind or visually impaired. Specifically, it seeks to
module to direct the user to nearby destined places. develop a portable Electronic Travel Aid (ETA) for
The system consists of a glass with camera and visually impaired users, along with the
sensors, an hardware and an audio system. It uses accompanying radio frequency identification
the MATLAB software to detect obstacles around (RFID) localization infrastructure used to equip
as well as traffic signal pattern. The technique buildings.
involved behind this is image acquisition, feature The paper discussed by NazliMohajeri et al, (2015)
extraction, feature matching, pattern recognition describes about how blind people face several
and template matching. This helps the blind people problems in their life, one of the most important
to detect the obstacles as well as the traffic signal one is detecting the obstacles when they are
and to cross the road. walking. In this research, we suggested a system
with two cameras placed on blind person's glasses
which takes images from different sides. By

2014
All Rights Reserved © 2016 IJARECE
 ISSN: 2278 – 909X
International Journal of Advanced Research in Electronics and Communication Engineering (IJARECE)
Volume 5, Issue 7, July 2016

comparing these two images, the obstacles would microcontroller part. The output from MATLAB is
be detected. given to the microcontroller. The microcontroller is
The paper as in [5] (Massimo Bertozzi and Alberto programmed to stimulate the output in the form of
Broggi ,1998) describes about a stereo vision-based audio commands using the audio playback and
hardware and software architecture to be used on record circuit. The ultrasonic sensors are used to
moving vehicles to increment road safety. Based on detect the distance between moving vehicles and to
the systems parallel hardware, it allows to detect intimate them.
both generic obstacles (without constraints on IV.IMPLEMENTATION METHOD
symmetry or shape) and the lane position in a A prototype was done where a model traffic
structured environment (with painted lane scene with a car and a bus was created. Templates
markings) at a rate of 10 Hz. are separately created in the MATLAB. The
The paper (Kirubakaran.S, 2001) as in [1] describes camera was connected to the MATLAB from
a smart system for visually impaired, that make use which a live video was captured and snapshots
of ultrasonic sensor and RF transceiver as assistive were taken from that live video. The snapshot is
devices. Visually impaired individuals find converted from RGB to gray scale image.
navigation difficult as they struggle everyday in Then that snapshot was compared with
performing actions for passing obstacles and templates stored using the concept of template
hurdles in their path. In order to help blind people matching. Template matching is a technique which
navigate safely and quickly this system is proposed. is used in digital image processing. It will identify
This system is based on embedded technology. small parts of the target image which matches with
III. BASIC BLOCK DIAGRAM the template image stored in MATLAB.Then
features of the template were surfed and circles
were plotted. This was based on feature extraction
method in MATLAB. Feature extraction is a
special form of dimensionality reduction. The main
aim of feature extraction is to get the most relevant
information from the original data and represent
that information in a lower dimensionality
space.Initial matches are done with the entire scene
image. Finally filtered matches are made which
exactly matches the template with the scene. Once
the matching is made, a message box is displayed
indicating that the obstacle is detected. The
MATLAB output was sent to the microcontroller
kit with max232 serially through a serial cable.
Themicrocontroller (AT89C51) was
Fig (a) block diagram programmed using keil-c software. For serial

The microcontroller used is AT89C51. The kit communication a Baud rate of 9600 was fixed and

includes MAX232 for serial communication. The the clock frequency was set to 11.0592 Hz. The

pc with MATLAB software is connected to the microcontroller on receiving output from

2015
All Rights Reserved © 2016 IJARECE
 ISSN: 2278 – 909X
International Journal of Advanced Research in Electronics and Communication Engineering (IJARECE)
Volume 5, Issue 7, July 2016

MATLAB in turn gives commands to the

APR33A3 kit. The APR consists of eight channels
and three modes of operation. The audio playback
and record kit gives audio commands to the user.
V. SIMULATION OUTPUTS
The fig (b) displays the target bus which is
extracted from the scene image. This target bus is
saved as a template in the MATLAB. The fig (c) Fig (d) surfing features of template
shows the model of traffic scene image which was
captured from the live video. The scene was
created using a model car and model bus. The
snapshot of this image is used as the target image.
The fig (d) describes the process of surfing the
features of the bus. The circles are plotted on the
bus. The fig (e) displays the initial matches. The
circles which are plotted on the bus will be
matched with the traffic scene image. The fig (f)
Fig (e) initial matches
shows the process of final matching. In this final
matching only the particular template will be
matched with the scene image. After the template
has been matched with the scene image a dialog
box will appear showing that the particular
template is detected. The fig (g) shows the message
box indicating the detected template.

Fig (f) final matches

Fig (b) bus template

Fig (c) scene image Fig (g) message box

2016
All Rights Reserved © 2016 IJARECE
 ISSN: 2278 – 909X
International Journal of Advanced Research in Electronics and Communication Engineering (IJARECE)
Volume 5, Issue 7, July 2016

VI. PROTOTYPE OF WORKING MODEL moving vehicles on the road and traffic signal and
The working model consists of the MATLAB intimating the user in the form of audio commands.
software, AT89C51 microcontroller and audio The future expansion would be implementing GPS
playback/ recorder. The image processing module in the system so that if the user tells the
technique is used in MATLAB. The simulated destination , using voice recognition it would detect
output is sent into the microcontroller. The the place and give commands to the destined
microcontroller development kit consists of places.
MAX232 which is used as a level converter. Keil REFERENCES
micro vision is used for loading the serial [1]. S.Kirubakaran,smart guidance for blind

communication program into the microcontroller. people(2011),current trends in information technology.

[2]. G.Balakrishnan, G.Sainarayanan, wearable stereo vision
When the simulated output sends a character from
for blindly impaired people(2012), mechatronics&
the MATLAB it will be sent to the microcontroller automation(ICMA) international conference.
through serial communication. For serial [3]. J.Choi, crosswalk and traffic light(2013)frontier of

communication the registers TMOD, SCON, TH computer vision workshop.

[4]. Nazilmohajeri, Roozbehraste, obstacle detection
and TR are set. Then microcontroller gives
(2007),proceedings of world conference vol 1:2007,P687.
commands to the audio playback/recorder. In APR [5]. Massimo Bertozzi,AlbertBroggi, a parallel real time
voice is recorded in different channels. According system for generic obstacle and lane(2001), IEEE

to the output received from controller it gives audio transactions on image processing
[6]. B.L.Meijer. experimental system for auditory image
commands. Fig(h) shows the prototype of the
representation (1992), IEEE transaction on biomedical
working model. engineering.
[7]. F.Hong, A.Chekima, ETA for visually impaired(2011),
international journal of disaster recovery and business
connectivity, IEEE a new vision vol.3,aug.

R.Mohanapriya, Department of ECE, S.A.Engineeringcollege,

Chennai, India, 8608028808.
U.Nirmala, Department of ECE, S.A.Engineeringcollege,
Chennai, India, 9840988863.
C.Pearlin Priscilla, Department of ECE,
S.A.Engineeringcollege, Chennai, India, 9003166368.

Fig(h) prototype of working model

VII. CONCLUSION
With the proposed system, if developed with more
accuracy, the blind people will able to move from
one place to another place without others help. If
such a system is developed, it will act as a platform
for the generation of much more features later. The
developed prototype gives good result in detecting

2017
All Rights Reserved © 2016 IJARECE