ISSN 2278-3091

Volume
Daniel Sekyere – Asiedu et al., International Journal 10, No.3,
of Advanced TrendsMay - JuneScience
in Computer 2021 and Engineering, 10(3), May - June 2021, 1646 – 1655
International Journal of Advanced Trends in Computer Science and Engineering
Available Online at http://www.warse.org/IJATCSE/static/pdf/file/ijatcse231032021.pdf
https://doi.org/10.30534/ijatcse/2021/231032021

The Review of Raspberry Pi Based - Systems to Assist the

Disabled Persons
Daniel Sekyere – Asiedu 1, Fezile Ozdamli 2
1
Department of Computer Information Systems, Near East University, 99138 Nicosia, Cyprus,
[email protected] https://orcid.org/0000-0002-9721-2629
2
2Computer Information Systems Research and Technology Centre, Near East University, 99138 Nicosia, Cyprus
[email protected] https://orcid.org/0000-0002-6699-3418

 reemphasized by [2]. Many people around the world are

ABSTRACT however incapable of vision. Although technology is
currently growing at a fast pace, people living with these
As people, eyesight is one of the most significant and disabilities cannot benefit from these technologies. Blindness
depended upon in acquainting ourselves with where we are can be described as a disorder in which the person is unable to
and what is around us. Freedom in exploring our new see or has no sense of light. In other words, it is a condition in
environment has to do with our capacity to survey these new which a person is forced to utilize his or her other senses for
environments, distinguish impediments as well as distinguish vision[3]. Thus, whether an individual suffers from a total or
between objects in these environments. This task as simple as partial vision deficiency, such a person is deemed visually
it may seem is one of the many problems encountered by the impaired.
blind in their daily lives. Which hinders their ability to freely
move about and carry out their day-to-day activities. Many Internationally, visual deficiency amounts to almost a billion
types of research have been carried out to develop systems to of the entire global population. Most of these individuals
assist the blind and the deaf. Some of these researches were either suffer from total or partial vision deficiency of which
developed with microcomputers as their central component. most are over the age of fifty years. Uncorrected refractive
This study seeks to investigate the literature for studies that defects and cataracts are the primary sources of vision
utilized raspberry pi to develop systems to assist the blind and deficiency and blindness according to a 2020 report [4]. The
the deaf. It as well attempts to define future research majority of blind people are poor and have low incomes or in
perspectives on the subject by addressing the research some cases no income at all. Most of them are found in Africa
questions stated in this study. The finding of this study would and Asia. According to new estimates, vision impairs are
enlighten researchers to be guided for future researches, as projected to rise to 115 million by 2050 [4].
well as enable them to see the flaws in the area of study in The challenges the blind and deaf encounter daily in
other to solve such problems. carrying out their activities has necessitated the inventions
of systems to ease their burdens. With the improvement in
Key words: visually impaired, Raspberry pi, Home technology in recent times, researchers have tried to develop
automation, Smart devices, object detection systems to alleviate their challenges. Shiyam (2016) in their
study developed a system that included an IoT module
1. INTRODUCTION comprising of accelerometer sensors and raspberry pi. The
As people, eyesight is one of the most significant and sensors capture real-time data related to the hand gestures of
depended upon in acquainting ourselves with where we are the user. Raspberry pi is used as a computing module that
what is around us. Freedom in exploring our new transfers data to the cloud and is embedded with Python
environment has to do with our capacity to survey these new Script with functionality to produce the desired output. The
environments, distinguish impediments as well as distinguish output is expected to be an E-mail and voice output. The
between objects in these environments. This task as simple as accelerometer sends the data to raspberry pi in support of
it may seem is one of the many problems encountered by the the Wi-Fi module. The Python Script on raspberry pi has the
blind in their daily lives. Which hinders their ability to freely adxl libraries to accommodate and understand the sensor’s
move about and carry out their day-to-day activities [1]. data. Raspberry pi 3 controllers are used for interface and
Having eyes to acquire and relay sensory information to the image processing purposes. The Voice recognition module
brain is a critical feature of human physiology. Also, 83% of is used for converting audio to text with the help of models
awareness of our surroundings is accessed through the eyes of neural networks that convert text into various sign
and therefore it is essential for human survival as was gestures. In this way, complete communication takes place
on both sides.

1646
Daniel Sekyere – Asiedu et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 1646 – 1655

Sign Language, is the basic form of gesture for the deaf and develop such systems for both the users and the developers.
dumb. By using sign language, they can convey their Furthermore, for researchers to know what has been done so
message. Bhagat-Thakre (2020) said the existing system far and the directions in which future research should be
has various disadvantages which are listed below. The encouraged? Additionally, the study seeks to investigate the
researcher developed a Braille system and disclosed that literature for studies that utilized raspberry pi to develop
harm can be induced to the surface. Such inductions could systems to assist the blind and the deaf. This study seeks to
be coursed by writing with a spot of indelible ink on the define future research perspectives on the subject by
surface of the system, as well when a blind person injures addressing the questions below.
the fingers, it would be difficult for such person to read from
the system. The braille device is therefore more prone to RQ1 What are the advantages, disadvantages of using
problems and inhibits reading. The main defect in the raspberry pi systems for disabled users?
Finger reader is that it can only process the English
Q2 What are the advantages, disadvantages of using
language and could not able make out for the deaf and dumb
raspberry pi systems for developers?
Sign language is frequently used, but it cannot be easily
understandable by those who do not understand the RQ3 What are the limitations of the current research on
language. raspberry pi projects?
Raspberry pi 3B+ is a less expensive computer but with a
sizeable amount of memory space with the size as small as a 2. METHODOLOGY
credit card. It utilizes USB and HDMI to connect to
keyboards, mouse, and monitors to operate the input-output PRISMA is one of the commonly used systematic literature
device of the device. It is easy to operate and program. One review methods in most academic research. This method of
doesn't need a deep knowledge of it to use. Raspberry pi acts literature review guides the researcher in acquiring
as a controller to control and monitor various operations knowledge especially in terms of previous research, makes the
like the processing of images and generating the research gaps more visible, and guides the researcher on how
appropriate words. Raspberry pi takes as input images of to improve existing knowledge in the current area of research.
hand gestures or various signs from a camera and compare A literature review has to be systematic and must follow a
the input to a stored database and generate appropriate text routine analysis for it to be exact or close to being exact [12].
[7]. This research is therefore consistent with the application of a
In recent years, many new solutions have been designed for Preferred Reporting Items for Systematic Reviews and
the elderly, the physically impaired, such as the blind and Meta-Analyses (PRISMA) method [13], [14].
the deaf as well as patients living on their own to give them
a sense of independence. Controlling unautomated devices As a method of reviewing the literature, PRISMA helps
may be a challenging situation popular among disabled researchers to make guided decisions with regards to the
persons [8]. An IoT or smart home is interconnected to selection of their research questions, as well as the selection of
devices and sensors which provide information on home their variables. PRISMA also helps the researchers in the
activities and help users monitor these devices. These selection of the right data that would help them to make
homes are fitted with cameras, actuators, and wireless guided research [15].
networks to establish communication between the smart
home and the devices [8]. A great way of achieving IoT is to With the help of meta-analysis and the PRISMA method of
utilize speech recognition to control and coordinate devices collecting data, this research intends to acquire resources that
and sensors. Many developers in their studies have utilized would be beneficial in developing a system with the raspberry
microphones for connectivity in smart homes and while pi to aid visually impaired and deaf persons [15].
some others have added various already made products to
submit speech commands to their systems for the same
2.1 Data collection and search process
purpose [9], [10]. Jain, Nandakumar and Ross (2016) in
their study, however, emphasized that developed systems
The study utilized two databases, which are Web of Science
that use speech recognitions have various challenge such as
(WOS) and Scopus because both databases are rich with the
proximity, security as well as noise which ultimately affect
information required by researchers. The knowledge from
the accuracy of these systems.
these databases covers scholarly articles that support one
Despite developments in this area of studies, it was found another in material, method, and strategies in most cases.
that no systematic literature review carried out to assess and
investigate the literature for studies that utilized raspberry
pi to develop systems to assist the blind and the deaf. This
has been a major gap for current developers to assess the
advantages and disadvantages of the use of raspberry pi to

1647
Daniel Sekyere – Asiedu et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 1646 – 1655

Table 1: Data collection and description texts were excluded.

 Studies outside the
Data collection Description period of 2016 to 2020.
 Studies that are not
Database Journals and articles were peer-reviewed journals.
obtained from Web of
Science (WOS) and Scopus 2.2 Alternative use of the keywords
Search string SCOPUS:
TITLE-ABS-KEY The search for the study was conducted based on the
((raspberry AND pi) AND following approach to answer the research questions in this
( "disabled" OR "deaf" manner. Alternative use of the keywords as shown in Table 1
OR "visually impaired" were used to collect relevant studies on the research question.
OR "physically disabled" ) Concatenating those alternative words, the search strings
) were formulated and were used to search the various
WOS: ALL FIELDS: databases
(raspberry pi) AND
TOPIC: (disabled OR deaf Table 2: Alternative use of the keywords
OR visually impaired OR
physically disabled) ) Blind/deaf Raspberry pi
Publication dates The research is restricted to Blind Raspberry pi
articles and journals visually impaired
published between 2016 physically challenged
and 2020 deaf
Language of publications The research is restricted to
articles and journals The string was however slightly modified when used on the
published in English individual online databases according to their interface
Included criteria  The research was requirements while keeping the logical order consistent for
focused on Raspberry preliminary searches. Full-text journal articles will be
pi projects to assist the examined to select those to be included in the study and the
blind and deaf. articles which will not meet the inclusion criteria were
 The research was excluded.
focused on studies
published from 2016 to 2.3 Study selection criteria
2020.
 The research was When the findings have been reached, the filtering criteria are
focused on studies used to screen out the unrelated studies. Papers retrieved from
published only in SCOPUS and WOS were independently sorted for inclusion
peer-reviewed and excluding criteria, by the two researchers. Paper was
journals/Articles. considered only if two or all the experts approved its
 Only English Language inclusion. 74 articles were obtained from the basic search
articles were totally. Out of that number, 13 were excluded due to
considered. duplication of these articles for the two databases. The use of
 Articles with a special raspberry pi to develop systems to assist the visually impaired
focus on raspberry pi and deaf was the focus of this study, therefore studies that
projects to assist the were not exactly related to this focus were excluded upon
blind and the deaf in reading the titles, abstracts, and keywords. Consequently, 17
tittles and keywords of the articles were further excluded resulting in 44 articles
were considered. for this study. After both researchers examined the suitability
 Only articles were of the full texts separately, 20 articles that had access to the
considered in this full text and fit the purpose were selected for in-depth
study. analysis.
Excluded criteria  Studies developed using
systems other forms of
microcomputer.
 Studies that are not
published with full

1648
Daniel Sekyere – Asiedu et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 1646 – 1655

(2020) navigation detect objects this method as it

Number of articles Number of articles system for and navigating removes
retrieved from retrieved from blind people shadows. This
database searching database searching using aids the blind to
(SCOPUS and WEB other sources raspberry-P move freely
OF SCIENCE) I”
(n = 0)
“An To improve Raspberry Pi
(n = 74)
Bin architecture existing can carry out
Shahriman et improveme architecture projected
al., (2019) nt of with modern functions of e
Articles after duplicates removed (n = 61) electronic technology Braille Quran.
Braille which is the There is a
Quran Raspberry Pi reduction in
based on 3B+ battery
Number of articles Number of articles Raspberry consumption as
retrieved after excluded (n = 17) Pi” well by 1.2%
keywords and title /hour
screened (n = 44)
Vasanth, “A Self To assist mild The system can
Macharla & Assistive deaf and blind use network
Full-text articles Full-text articles Varatharajan Device for persons to protocols with
assessed for excluded, with (2019) Deaf & work the help of
eligibility after reasons (n = 24) Blind successfully in Google API
abstract screened People their servers that
(n = 21) Using IOT” surroundings provides
affordable aids
for people living
with impairment
Studies Latha et al., “A smart To provide a Findings show
included (2019) reader for smart bot that that the smart
analysis blind will aid read for the blind
(n =20) people” visually has a positive
impaired outcome as the
persons. openCV is easier
Figure 1: Figure 1. Flow diagram of the publication selection to use in
process comparison to
the PC platform.
It aids in
accessing data in
3. SOME COMMON MISTAKES different forms.
Rajyashree, “Communi To attain the The model
This paper reviews past literature to determine the advantages et al. (2019) cation finest enables the deaf,
assistant for technology dumb and blind
and disadvantages of raspberry pi to develop systems to assist the deaf, that is directed to interact with
disabled people for users and developers. The table below dumb and at aiding the their
blind” blind to hear surroundings
summarizes the studies from each of the 20 papers used in this the text that is through the use
research. converted to of a handy
sound. device that gets
rid of all
Table 3: Summary of Related Studies interferences in
communicating.
Research Research Aim of Study Findings This helps in
title developing
Zubov “A Smart To use the The new them. The
(2018) City smart city improved Voila- system is
Assistive assistive Jones face powered by
Infrastructu structure sensor with other Raspberry Pi to
re for the obtainable for features such as support the deaf,
Blind and B &VI to aid eye and nose dumb and blind.
Visually interaction was Sahoo, Lin & “Design To provide an The lifestyle of
Impaired with recommended to Chang and aid that will VCPs will be
People: A additional aid in the image (2019) implementa assist with improved
Thin Client subsystems processing tion of a obstacle drastically as
Concept” through IoT speed. This can walking detections they will be able
procedures only be applied stick aid for to live on their
with the use of visually own with the use
HOG+EDC challenged of this device
since the rate of people” thereby making
detection is not them
100% self-assured in
Anandan, “Advanced The provision There is a good the environs,
Manikandan indoor and of assistance detection result they find
& Karthick outdoor for the blind to with the use of themselves.

1649
Daniel Sekyere – Asiedu et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 1646 – 1655

The use of the optical use of artificial shopping

APP as a character intelligence through the use
tracking device recognition and OCR of L293Ddriver,
for parents and ” optical character
also help in recognition, and
emergencies. Raspberry Pi.
This gives both Hapsari, “Face To provide a The device
the user and Mutiara & recognition smart cane serves as a tool
parent some Tarigan smart cane that has facial aid that will
form of security (2019) using recognition assist the blind
Sharath, “Developm To enable This system has haar-like function with to recognize
Muthangi & ent of facial a high accuracy features and the use of other people that
Premananda self-assisted recognition rate of face eigenfaces” Haar-like have a distance
(2020) voice and text recognition features and of 0.25 to 1.5
module for apprehensions where blind Eigenfaces meters. This
visually by the use of people are smart cane has a
impaired” Local Binary unable to see 91.67%
Pattern properly. As accuracy if the
Histogram well as read tests face to recognize
(LBPH) accurately. is a straight
The LBPH position to the
system is camera
portable and can Zeineddine et “Low Cost The provision The findings
be carried al (2020) Electronic of a showed that the
anywhere by Brailler” budget-friendl design is low
visually y cost and
impaired people. LCE Brailler effective and
The system can will vocalize users were
be improved and convert satisfied with it.
upon to include typed No errors were
other languages characters to experienced is
as well as alphabetical using the control
navigation characters. buttons which
systems. To aid in worked
Ismail A., "Developm To provide the The system was interactions accordingly.
Abdlerazek ent of smart aged and able to assist among
S., healthcare people living patients and the students,
El-Henawy system with disability aged to have teachers, and
(2020) Based on to have speech control over IoT parents when
speech recognition devices, through learning.
recognition that is a their Akour et al “Mobile To develop a Their research
using budget-friendl compatibility to (2020) Voice system that was able to
support y system and aid in speech Recognition uses an detect the vocal
vector can have recognition. It Based for Android command with
machine access to the provided Smart mobile phone high accuracy. It
and internet easily designed to Home to control with mentioning
dynamic to be used in provide privacy Automation home that their system
time homes and in managing Control” appliances was cheap to
warping” hospitals. patient devices with a voice develop and
in hospitals and command proposed that
at homes for the from the user. future research
elderly. There could be made to
was a 97% level include the
of accuracy system helping
higher than the elderly and
previous disabled persons
systems that as well.
were 79% by the Baskaran & “Real Time To develop a Their system
use of DTW Balachander Object system to could accurately
with SVM that (2019) Detection would assist in detect objects
was able to System” locating with the use of a
differentiate misplaced high pixel
sound segments. items. As well camera and
There is a as assist the could as well
system that is visually adjust to new
unable to detect impaired with situations
voices that are face because of the
not clear. recognition. learning
Meera, “Effective To make The system can algorithms used.
Sharmikha shopping movement in provide smart Moreover, their
Sree & method for an urban shopping for the system could be
vvalarmathi visually shopping blind with no controlled
(2019) impaired environment form of remotely as well.
people easy for the assistance. This Prabhu, “Smart To develop a They found from
using blind with the facilitates Ruban Glass for smart glass to literature and

1650
Daniel Sekyere – Asiedu et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 1646 – 1655

Thomas, Visual assist the their project size of an A4

Senthil Impaired visually that, paper.
Kumar People” imparted, the sound-based The developed
(2019) deaf and the assistance system did not
dumb. device was most require internet
suitable for the connectivity to
visually operate. Their
impaired. They proposed system
as well utilized a can be used to
servo engine to assist the editing
produced touch eBooks and web
for the deaf and pages.
dumped person. Dunai et al “Euro To develop a The researcher
The researcher (2017) Banknote handy system found out that,
stresses that their Recognition to enable the due to the
developed System for blind to infrared laser
system was Blind identify and camera, their
affordable to People” differentiate system was able
produce. Euro notes to detect
Latha et al. “A smart To develop a Their system banknote even
(2019) reader for system, assist was able to able with less
blind visually to inform the illuminated
people” impaired to visually environments.
read in impaired to read They as well
developing the text and used algorithms
countries proposed to that could be
eliminate the trained which
limitations of made their
their current system more
system. effective. They
Habib et al. “Staircase To detect a The developed however
(2019) Detection to staircase and system could realized
Guide ground by detect the difficulties in
Visually utilizing staircase with identifying
Impaired already high accuracy in crumpled
People: A trained models the illumined banknotes. In
Hybrid and ultrasound environment their study, it
Approach” sensors using RGBD took a relatively
cameras and long time to
with higher train the
accuracy with algorithms,
the use of whereas it took
ultrasonic 11 seconds to
sensors. They recognize an
however already
mentioned a recognized note.
future They intended to
development of improve the
the system into a algorithms for
more accuracy in note
user-friendly detection in their
device for future
industrial researches and
production as well to be able
would be to detect fake
necessary. banknotes. They
They also also intended to
mentioned their utilize memory
system would be usage
expanded to judiciously in
detect other their future
forms of research.
obstacles in the Rajbongshi et “Bangla To extract Accuracy of
way of the al. (2020) Optical Bangla text 97% was
visually Character from Bangla attained in the
impaired. Recognition images and image to text
Venkateswarl “Text to To develop a Their developed and further change conversion since
u et al (2017) speech cheap system system was able Text-to-Spe the extracted the image
conversion” to help users to to convert text to ech text to speech capture range of
listen to the speech with an Conversion raspberry pi is
content of a average using not of high
text and image processing Raspberry quality. And was
period of not Pi” almost
more than 3 impossible to
minutes for the capture images

1651
Daniel Sekyere – Asiedu et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 1646 – 1655

at night. They disabled persons. The literature revealed that it is affordable

also found noise
in the speech due and advantageous for the disabled person when raspberry pi is
to algorithms used to develop systems for such persons.
used for the text
to speech
conversion. The Akour et al (2020) wrote about the comfort users attain for a
researcher, home automated system build with raspberry pi to control
therefore,
decided to their home appliances with a voice command. They as well
enhance the said it is beneficial to use this device to develop a smart
algorithms, you lighting system in a museum by combining it with smart
hight quality pi
cameras, and thermal sensors to feel the presence of tourists. In their view,
expand the most of the recently developed systems with this device are
research to the
capture of
robust and able to withstand the variance in accents in
handwritten languages and noise. These systems are easy to use thereby
scripts in their being very helpful and accessible to the blind, disabled
future study.
persons as well as the elderly. they are additionally secured by
being able to validate and verify the voice of owners of the
3.2 Advantages and disadvantages of using raspberry pi
systems for disabled users systems. A monotonous task performed by humans can easily
be replaced by systems developed with raspberry pi. They
Throughout literature, most authors have mentioned the further mentioned that activities carried out in dangerous
advantages of using raspberry pi to develop systems to assist environments by humans can safely be replaced by these
the blind in various ways. Zubov (2018) in his study found the systems. They also estimated a low cost of systems produced
raspberry pi-based system to be very useful to visually with raspberry pi. In their view, it would be cheap but accurate
impaired persons, in the sense that such systems are cheap to of the functions the systems are designed for, there would easy
develop. The author as well mentioned the portability of a for disabled persons in third world countries to afford.
raspberry-based system for easy carriage of visually impaired
persons. Venkateswarlu et al (2017) also mentioned the Habib et al (2019) also complain about the low quality of the
portability of the device and how it makes the systems image produced by the raspberry camera during their test.
portable to be easily carried around by the blind, as well as it This is their observation affected the overall accuracy of the
cheap and it not depending on the internet making it performance of the entire system during testing. They
comfortable to be used anywhere by the blind without the however mentioned that it was advantageous using a
internet connectivity. Additionally, Hapsari, Mutiara & raspberry pi to develop a real-time image processing system
Tarigan (2019) spoke about the portability of the raspberry for identification. Hapsari, Mutiara & Tarigan (2019) in their
pi-based system. The fact that raspberry pi is a credit card size study mentioned that raspberry pi has been one the most
device with the capacities of a full computer makes it easy to popular devices used for face recognition for classroom
be used to implement IoT, an assistive and portable attendance checking, as well as the development of security
server-based system to help the blind, the deaf, and the doors.
elderly.
3.3 The advantages and disadvantages of using
In the production of prototypes and development of the raspberry pi systems for developers
assistive system, it is important to ensure the overall cost price
is not expensive. This it is because, most of the disabled Currently, there are several microcomputers on the market for
person go through financial challenges, especially those in the development of systems to assist the disabled person.
the third world countries. Dunai et al (2017) noted the need Some of the devices are UP2 Squared, Huawei HiKey 960,
for a cheap and simple assistive system for the blind because ODROID-XU4, and Arduino Uno just to mention a few.
of their constant use-dependent of such systems, therefore a However, most of them are either expensive or have a lower
raspberry pi-based system is appropriate for such designs. processing speed and fewer functionalities as compared to a
Venkateswarlu et al (2017) additionally in their study raspberry pi. Raspberry pi 3B+ is a less expensive computer
mentioned the efficiency and cost-effectiveness of developing but with a sizeable amount of memory space with its
a system based on a raspberry pi to convert text to speech to dimension as small as a credit card. It utilizes USB and HDMI
assist blind persons. Sharath, Muthangi & Premananda to connect to keyboards, mouse, and monitors to operate the
(2020) also emphasized the cost and security of raspberry pi input-output of the device. It is easy to operate and program.
as one of the main advantages of using it to develop a system One doesn't need a deep knowledge of it to use. Rajbongshi et
to assist blind persons. Because it ends up coursing the total al. (2020) reiterated in their study that, raspberry pi is a
cost price to be low thereby making it affordable for the poor Linux-based system, which makes it easier for project

1652
Daniel Sekyere – Asiedu et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 1646 – 1655

development. They mentioned that most of the libraries for 3.4 Limitations of the current research on raspberry
text-to-speech conversions are compatible and easily installed pi projects
on Linux systems. In their study, they emphasized the use of
high pixel cameras to intensify the accuracy of text captured Despite the researches carried out in this area of study, there
by the raspberry camera, in their opinion has been a still some setbacks in the development of these systems.
limitation. The researcher also mentioned the use of advanced Rajbongshi et al. (2020) observed that there was no system
text-to-speech algorithms in other to reduce noise during the developed to read the handwritten text, this they predicted
process. Latha (2019) said one of the advantages of using a should be an area future researches should be undertaken.
raspberry pi based system to assist the blind in banknote [31] Dunai et al (2017), for detection fake banknote for the
recognition is that it has a short processing time as compared visually impaired, there is a missing gap in arear of research
to other microcomputers. Dunai et al (2017), reiterated that and they proposed, this should be researched further in the
the less expensive nature of systems developed with raspberry future to assist visually impaired to be able to identify and
pi. They, however, wrote about the camera quality which differentiate fake banknotes as well as other documents.
brought a difference in the quality of the captured image in an Habib et al (2019) said systems that would be able to detect
illuminated area and a dark area. obstacles such as stationary and motional objects should be
developed to assist the blind. They further mentioned the need
Venkateswarlu et al (2017) in their study emphasized that to develop a single unit raspberry pi-based system. Whereas
raspberry pi has a fast-processing speed. It usually took research in the area of developing systems to assist the blind
approximately three minutes to convert text to speech when and deaf in other parts of the world is increasing, little is been
used in their project. They found it very tolerant with accuracy done in regions like the Middle East and Africa. Akour et al
above 98% in converting the text of an A4 sheet size (2020) mentioned that future researchers should invest time
document to audio. One major advantage they mentioned in into the development systems that would aid the deaf, system
their study was the fact a raspberry pi system can be a that would take prompts and commands from them.
standalone system; in that, it accurately works without .
internet connectivity. Akour et al (2020) described the 4. DISCUSSION and CONCLUSION
raspberry pi as a server to which other components can be
connected as a central point in the development of smart This study conducted systematic literature on developed
systems. Raspberry pi can be then connected to the cloud, this projects that were based on the raspberry pi to fundamentally
forms the core of the internet of things and smart automation assist the blind and the deaf. It considered the advantages and
and control of devices and sensors. On the side of the disadvantages of using raspberry pi to implement these
production of raspberry pi-based systems, they also estimated projects. It additionally considered the merits and demerits of
a low cost in the systems produced with raspberry pi. In their implementing projects based on these devices from the
view it would be cheap but accurate of the functions the perspective of the developers. Moreover, the research
systems are designed for, there would affordable for highlighted some noticeable missing gaps within these kinds
developers for experimental and implementational purposes. of projects. It was evident that most of the projects are carried
Baskaran & Balachander (2019) said the price of raspberry pi out by researchers from India. It is therefore strongly
makes it suitable for experimental purposes, developers recommended that academician from developing countries or
would not afraid to experiment with it because it is cheap as third world countries, as well as developing regions, should
compared to a personal computer have has the functionalities make more investigations into the subject matter in other to
of a personal computer. This they said would encourage enhance their knowledge in the subject area, improve on the
innovation and novelty in the development of systems to assist existing system and advance the currently used algorithms
the blind and the disabled as a whole with the aid of employed in the types of development. These would go a long
technology. The scalability of development is very essential, way to assist the blind and the deaf in such geographical areas
Ismail et al (2020) in their study mentioned that systems who are poor and cannot afford the expensive assistive
developed based on raspberry pi are easily scalable. Thus, products on the commercial market but have challenges in the
more sensors and devices can be added to the system for more autonomous execution of their regular activities.
functionalities as and when needed.
Conversely, Hapsari, Mutiara & Tarigan (2019) complained REFERENCES
about the processing speed of raspberry pi devices used to [1] D. Sekyere-Asiedu, “a Raspberry Pi Based System To
implement image processing systems. The image processing Help the Visually Impaired At Home,” 2018, [Online].
algorithms process involving and need very fast Available:
computational speed for such actives. They further criticize http://docs.neu.edu.tr/library/6712561709.pdf.
the memory space. [2] S. Venkateswarlu, D. B. K. Kamesh, J. K. R. Sastry,
and R. Rani, “Text to speech conversion,” Indian J.

1653
Daniel Sekyere – Asiedu et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 1646 – 1655

Sci. Technol., vol. 9, no. 38, pp. 10–12, 2016, doi: 10.26666/rmp.ijbm.2019.1.2.
10.17485/ijst/2016/v9i38/102967. [16] D. Zubov, “A Smart City Assistive Infrastructure for
[3] C. Jicol et al., “Efficiency of Sensory Substitution the Blind and Visually Impaired People: A Thin Client
Devices Alone and in Combination With Self-Motion Concept,” Brain-Broad Res. Artif. Intell. Neurosci.,
for Spatial Navigation in Sighted and Visually vol. 9, no. 4, pp. 25–37, 2018.
Impaired,” Front. Psychol., vol. 11, no. July, 2020, doi: [17] M. Anandan, M. Manikandan, and T. Karthick,
10.3389/fpsyg.2020.01443. “Advanced indoor and outdoor navigation system for
[4] “WHO | World Health Organization.” blind people using raspberry-PI,” J. Internet Technol.,
https://www.who.int/ (accessed Feb. 21, 2021). vol. 21, no. 1, pp. 183–195, 2020, doi:
[5] S. Raghul, M. Surendhar, K. Suresh, and R. 10.3966/160792642020012101017.
Hemalatha, “Raspberry-Pi Based Assistive Device for [18] A. S. Bin Shahriman, S. Bin Mad Saad, Z. A. Bin
Deaf, Dumb and Blind People,” Ijsrset, vol. 142, no. 2, Ahmad, M. Bin Hussein, M. A. Bin Mat Dzahir, and
pp. 167–174, 2016. M. A. Bin Mat Dzahir, “An architecture improvement
[6] M. S. Journal, “Smart interaction system for blind and of electronic Braille Quran based on Raspberry Pi,”
dumb,” vol. IX, no. Vi, pp. 2270–2274, 2020. Int. J. Innov. Technol. Explor. Eng., vol. 8, no. 11
[7] A. Dayal, N. Paluru, L. R. Cenkeramaddi, S. J., and P. Special issue 2, pp. 395–399, 2019, doi:
K. Yalavarthy, “Design and Implementation of Deep 10.35940/ijitee.K1063.09811S219.
Learning Based Contactless Authentication System [19] K. Vasanth, M. Macharla, and R. Varatharajan, “A
Using Hand Gestures,” Electronics, vol. 10, no. 2, p. Self Assistive Device for Deaf & Blind People Using
182, 2021, doi: 10.3390/electronics10020182. IOT: Kathu-Kann Thaan Thunai Eyanthiram,” J.
[8] A. Ismail, S. Abdlerazek, and I. M. El-Henawy, Med. Syst., vol. 43, no. 4, 2019, doi:
“Development of smart healthcare system based on 10.1007/s10916-019-1201-0.
speech recognition using support vector machine and [20] L. Latha, V. Geethani, M. Divyadharshini, and P.
dynamic time warping,” Sustain., vol. 12, no. 6, 2020, Thangam, “A smart reader for blind people,” Int. J.
doi: 10.3390/su12062403. Eng. Adv. Technol., vol. 8, no. 6 Special Issue 3, pp.
[9] F. Pisani, F. M. C. De Oliveira, E. S. Gama, R. 1566–1568, 2019, doi:
Immich, L. F. Bittencourt, and E. Borin, “Fog 10.35940/ijeat.F1285.0986S319.
Computing on Constrained Devices: Paving the Way [21] Rajyashree, O. Deepak, N. Rengaswamy, and K. S.
for the Future IoT,” Adv. Parallel Comput., vol. 35, no. Vishal, “Communication assistant for deaf, dumb and
June, pp. 22–60, 2020, doi: 10.3233/APC200003. blind,” Int. J. Recent Technol. Eng., vol. 8, no. 2
[10] K. AbualSaud et al., “A Survey on Mobile Special Issue 11, pp. 94–98, 2019, doi:
Crowd-Sensing and Its Applications in the IoT Era,” 10.35940/ijrte.B1015.0982S1119.
IEEE Access, vol. 7, pp. 3855–3881, 2019, doi: [22] N. Sahoo, H. W. Lin, and Y. H. Chang, “Design and
10.1109/ACCESS.2018.2885918. implementation of a walking stick aid for visually
[11] A. K. Jain, K. Nandakumar, and A. Ross, “50 years of challenged people,” Sensors (Switzerland), vol. 19, no.
biometric research: Accomplishments, challenges, and 1, 2019, doi: 10.3390/s19010130.
opportunities,” Pattern Recognition Letters, vol. 79. [23] S. Sharath, R. Muthangi, and B. S. Premananda,
pp. 80–105, 2016, doi: 10.1016/j.patrec.2015.12.013. “Development of self assisted voice module for visually
[12] Y. Xiao and M. Watson, “Guidance on Conducting a impaired,” Int. J. Emerg. Trends Eng. Res., vol. 8, no.
Systematic Literature Review,” J. Plan. Educ. Res., 9, pp. 6247–6253, 2020, doi:
vol. 39, no. 1, pp. 93–112, 2019, doi: 10.30534/ijeter/2020/216892020.
10.1177/0739456X17723971. [24] S. Meera, R. Sharmikha Sree, and K. valarmathi,
[13] M. Abelha, S. Fernandes, D. Mesquita, F. Seabra, and “Effective shopping method for visually impaired
A. T. Ferreira-Oliveira, “Graduate employability and people using optical character recognition,” Int. J.
competence development in higher education-A Eng. Adv. Technol., vol. 9, no. 1, pp. 5304–5306,
systematic literature review using PRISMA,” Sustain., 2019, doi: 10.35940/ijeat.A2961.109119.
vol. 12, no. 15, 2020, doi: 10.3390/SU12155900. [25] G. I. Hapsari, G. A. Mutiara, and H. Tarigan, “Face
[14] M. B. Nawaila, S. Kanbul, and F. Ozdamli, “A review recognition smart cane using haar-like features and
on the rights of children in the digital age,” Child. eigenfaces,” Telkomnika (Telecommunication
Youth Serv. Rev., vol. 94, no. July, pp. 390–409, 2018, Comput. Electron. Control., vol. 17, no. 2, pp.
doi: 10.1016/j.childyouth.2018.09.028. 973–980, 2019, doi:
[15] M. Nazri and M. Rudi, “Military leadership : A 10.12928/TELKOMNIKA.V17I2.11772.
systematic literature review of current research,” Int. J. [26] Z. Zeineddine, S. Sindian, N. A. H. Ahmad, and G.
Bus. Manag., vol. 3, no. 2, pp. 1–15, 2019, doi: Ismail, “Low cost electronic brailler,” WSEAS Trans.

1654
Daniel Sekyere – Asiedu et al., International Journal of Advanced Trends in Computer Science and Engineering, 10(3), May - June 2021, 1646 – 1655

Environ. Dev., vol. 16, pp. 811–819, 2020, doi:

10.37394/232015.2020.16.84.
[27] M. Akour, K. Al Radaideh, A. Shadaideh, and O.
Okour, “Mobile voice recognition based for smart
home automation control,” Int. J. Adv. Trends
Comput. Sci. Eng., vol. 9, no. 3, pp. 3788–3792, 2020,
doi: 10.30534/ijatcse/2020/196932020.
[28] T. S. H. Baskaran and B. Balachander, “Real time
object detection system,” Int. J. Eng. Adv. Technol.,
vol. 8, no. 6, pp. 59–62, 2019, doi:
10.35940/ijeat.E7440.088619.
[29] V. Prabhu, D. Ruban Thomas, and V. Senthil Kumar,
“Smart glass for visual impaired people,” Int. J. Innov.
Technol. Explor. Eng., vol. 8, no. 12, pp. 3411–3415,
2019, doi: 10.35940/ijitee.L2581.1081219.
[30] A. Habib, M. M. Islam, M. N. Kabir, M. B. Mredul,
and M. Hasan, “Staircase detection to guide visually
impaired people: A hybrid approach,” Rev.
d’Intelligence Artif., vol. 33, no. 5, pp. 327–334, 2019,
doi: 10.18280/ria.330501.
[31] L. D. Dunai, M. C. Pérez, G. Peris-Fajarnés, and I. L.
Lengua, “Euro banknote recognition system for blind
people,” Sensors (Switzerland), vol. 17, no. 1, pp.
1–14, 2017, doi: 10.3390/s17010184.
[32] A. Rajbongshi, M. I. Islam, M. M. Rahman, A.
Majumder, M. E. Islam, and A. A. Biswas, “Bangla
optical character recognition and text-to-speech
conversion using raspberry Pi,” Int. J. Adv. Comput.
Sci. Appl., vol. 11, no. 6, pp. 274–278, 2020, doi:
10.14569/IJACSA.2020.0110636.

1655