International Journal of Research in Engineering, Science and Management 21

Volume-3, Issue-4, April-2020

www.ijresm.com | ISSN (Online): 2581-5792

Design and Implementation of Women Safety

System Using Mobile Application in Real-Time
Environment
S. Juhitha1, M. Pavithra2, E. Archana3
1,2
Student, Department of Computer Science & Engineering, Panimalar Institute of Technology, Chennai, India
3
Assistant Professor, Department of Computer Science & Engineering, Panimalar Institute of Technology,
Chennai, India

Abstract: Safety of Women has become a major problem in acceleration-based detection is that the most generally used
today’s World. The crime rates in case of women have risen to a technique, as current high-level devices integrate the
great extent. In order to reduce this crime rate and ensure women acceleration sensors.
safety a secure Mobile application is proposed. This application
requires an initial registration along with Emergency contacts. The planned system has 3 central components: sensing,
When the user is travelling from one place to another, the dynamic learning, and alerting. within the within the, we have a tendency
GPS tracking offered by G-Maps API is turned on to view the to take the benefits of the information base-based approach to
user’s location on a map. During an emergency situation, the gather realistic fall data. That is, we have a tendency to collect
victim can either shake the mobile phone to a specific frequency or the important measuring device information from the mobile
long press lower volume button. Immediately after this action an embedded sensors and record the corresponding user behavior
alert message containing the name of the victim, GPS location and
a help message is sent via SMS to all the registered Emergency to work out the specified parameters. within the second
Contacts and a Call is made to the Master contact. Suppose if the element, the planned system learns the link between the autumn
emergency contact person has the same application they can behavior and also the collected information. within the third
directly monitor through the dynamic GPS tracking system, element, the mobile phones alert pre- configured emergency
otherwise they can access the location through the message link. contacts through message. The experiment any investigates the
The Person can immediately rush to the spot to help the victim.
impact of various locations wherever the phone hooked up,
Suppose if the victim is in motion, the live location will be updated
for every certain seconds. Not only during Emergency situations, together with chest, waist, and thigh. Finally, we have a
even when the battery drains out, an alert message with the last tendency to conduct in depth experiments to gauge the
updated location is sent to the Emergency Contacts. additional power consumption ensuing from the autumn
detection software package.
Keywords: G-Maps API, Alert message, Emergency Contacts,
SMS, GPS tracking system, Mobile application, Android, Master 2. Literature survey
contact, Sensor.
C. Wang says, falls are the number one cause of injuries in
1. Introduction the elderly. A wearable fall detector can automatically detect
the occurrence of a fall and alert a caregiver or a medical rescue
Today’s smartphones or mobile phones function the central
computing and communication device in people’s lives. These group for immediate assistance, mitigating fall-related injuries.
However, most studies on fall detection to date have focused on
phones engraft varied sensors, together with measuring device,
the accuracy of detection while neglecting power efficiency and
digital compass, GPS, and camera, enabling new applications
in varied domains like health care, social networks, and battery life, and hence the developed fall detectors usually
cannot operate for a long period (a year or more) without
environmental observation.
recharging or replacing their batteries. This fall detector reduces
This can be as a result of falls square measure a significant
health risk for older folks, decreasing the standard of life or its power consumption through both hardware- and firmware-
based approaches. This study also incorporates several human
perhaps leading to death. The autumn detection system will be
trials to develop and evaluate the device, including simulated
will be four categories: image-based systems, database-based
falls and activities of daily living. The LPFD reduces its power
approach, context-aware techniques and acceleration-based
detection. for instance, the advantage of image-based associated consumption using both hardware and firmware based
approaches. The parameters of the classifier were optimized to
context-aware approaches is a correct detection rate; the
information-based approach stores numerous detected user achieve high but balanced sensitivity and specificity using PSO
behavior into a database for numerous activities. The in the training stage. A benchtop power measurement test is also
 International Journal of Research in Engineering, Science and Management 22
Volume-3, Issue-4, April-2020
www.ijresm.com | ISSN (Online): 2581-5792

conducted to estimate the battery life with data from a one- the place he fell down [8].
week free-living trial. These experiments show that the fall Mihail Popescu says quite one third of regarding thirty-eight
detector achieves high sensitivity (97.5% and 93.0%) and million adults sixty-five and older fall annually at intervals the
specificity (93.2% and 87.3%) on training and testing datasets, United States of America. To handle the on top of drawback we
whilst providing an estimated battery life of 664.9 days. The have a tendency to propose to develop Associate in Nursing
main drawback is that the LPFD is designed to work in an acoustic fall detection system(FADE); which will mechanically
indoor home environment [2]. signal a fall to the observance caregiver. As against several
F. Sposaro says, injuries due to falls are among the leading existent fall detection systems that require the monitored person
causes of hospitalization in elderly persons, often leading to a to wear devices like accelerometers or gyroscopes within the
rapid decline in quality of life or death. Rapid response can least times, our system is completely unnoticeable by not
improve the patient’s outcome, but this is often often lacking requiring any wearable devices. to scale back the warning rate,
when the injured person lives alone and therefore the nature of we have a tendency to use Associate in Nursing array of
the injury complicates calling for help. This paper presents an acoustic sensors to urge sound supply height data. The sound is
alert system for fall detection using common commercially taken under consideration a warning if it comes from a supply
available electronic devices to both detect the autumn and alert set at a height on top of two feet. victimisation the peak of the
authorities. It is proven that fall detection algorithm makes the sound, has drastically improved the warning rate. This
system highly reliable. It is Reliable and reduce the number of dedicated fall detection system relies on a linear array of audio
false positives mean greater adoption by emergency services. sensors. We tested our system during a pilot study that consisted
An Android-based smart phone is used with an integrated tri- of a collection of twenty-three falls performed by a stunt actor
axial accelerometer. Data from the accelerometer is evaluated throughout six sessions of regarding unit of time every (1.3
with several thresholds based algorithms and position data to hours in total). The actor was antecedently trained by our
figure out a fall. The threshold is adaptive supported user nursing collaborators to fall like Associate in Nursing old
provided parameters such as: height, weight, and level of person. the employment of height data reduced the warning
activity. The algorithm modify to unique movements that a hourly rate from thirty-two to five at a 100% fall detection rate.
phone experiences as against similar systems that require users Though the warning rate has been reduced, it's not fully
to mount accelerometers to their chest or trunk. If a fall is sense, corrected, Different array shapes(circular) are often thought-
then a notification is raised for the user’s response. If the user about for additional reduction of warning rate [5].
doesn't respond, the system alerts pre-specified social contacts J. Dai says falls are a serious health risk that diminish the
with an informational message via SMS. If a contact responds standard of life among elderly people. With the elderly
the system commits an audible notification, automatically population surging, especially with aging "baby boomers", fall
connects, and enables the speakerphone. Certain interactions detection becomes increase. However, existing commercial
such as violently answering then ending a call or dropping the products and academic solutions struggle to realize pervasive
phone can break the Thresholds can change positions. If a fall detection. we propose utilizing mobile phones as a platform
contact confirms a fall, an emergency service will be alerted. for pervasive fall detection system development. To our
Our system provides a realizable, cost effective solution to fall knowledge, we are the first to do so. We design a detection
detection employing a simple graphical interface while not algorithm supported mobile platforms. We propose PerFaUD,
overwhelming the user with uncomfortable sensors [4]. a pervasive fall detection system implemented on mobile
N. Noury says that the main purpose is to automatically phones. The detection algorithm is designed based on mobile
detect the autumn of the person from one device, worn on the phone platforms. Experimental results show that PerFallD
trunk within the “para sagittal” plane in real time. The device achieves good detection performance and power efficiency. We
includes 2 accelerometers, an 8 hits’ RISC microcontroller, a implement a prototype system on the Android Gl phone and
buzzer and B push button. Good Results are obtained with conduct experiments to gauge our system. In particular, we
limited population, each fall was simulated, and all the compare PerFaIID's performance thereupon of existing work
participants were young people. On improving the decision and a billboard product. PerFailD show that it achieves strong
algorithm, interesting simulation results with more elaborated detection performance and power efficiency. PerFallD can be
data signal processing, combining a decision on fuzzy data and enhanced by integration with some extra protection devices,
a combination of decisions It is serially linked to a RF-modem. e.g., airbag based fall protector to reduce fall impacts and
The test was taken on 10 healthy young persons in different prevent fall related injuries [3].
situations and exhibits good specificity and sensibility.
Improvements can he made on the decision algorithm using 3. Existing system
nonstandard theories. The common alarm systems are not The major drawback with existing business product and
satisfactory: a loss of consciousness or a faint are not detected, tutorial analysis is that they need they need hinder pervasive fall
people suffering from the Alzheimer disease doesn’t even have detection. the bottom should be put in somewhere inside and
the intellectual ability to launch an expert to call for help from also the moveable device should be hooked up to a belt at the
 International Journal of Research in Engineering, Science and Management 23
Volume-3, Issue-4, April-2020
www.ijresm.com | ISSN (Online): 2581-5792

waist. Once the bottom receives the signal from the device style AN interface that permits users to manually disable the
indicating a fall, it will mechanically communicate with a tuned in to avoid false positive and to scale back transmission
planned emergency contact victimization the victimization prices. GPS tracking feature tracks the user lively when you are
phone. However, the most distance between the device and also the move after triggering the emergency button. When the
the base is proscribed. Fall detection will solely be conducted battery is running low, it automatically sends the location the
inside a little indoor atmosphere and older individuals might pre-stored contacts. This device works without internet
simply forget to bring the device with them, because it is an connectivity.
additional device that they rarely use in everyday life. what is 1) Sensing
more, these product area units overpriced. just in case of falls, The measuring device embedded in mobile phones performs
the event of automatic, reliable, and prompt fall detection the sensing. Associate in Nursing measuring device detects
systems plays a significant to ensure immediate help and acceleration, vibration, and tilt to see movement and precise
facilitate. orientation on the 3 axes. Apps use this smartphone device to
The Drawbacks of the system is when emergency situation see whether or not your phone is in portrait or landscape
occurs the user cannot protect the smartphones and they cannot orientation. It can even tell if your phone screen is facing
track the phones if they lost it. The user cannot be set the alert upward or downward. The measuring device can even notice
function when they are in risk situation and immediately they how briskly your phone is taking possession any linear
cannot pass their location to friends and family members to direction. once the person shakes the mobile at the next
intimate they are in risk. The internet cost is too high and time frequency, this measuring device helps in sensing the
delay to find out the IMEI number of mobile to track it. movement by victimization the x, y, z coordinates.

Fig. 2. Accelerometer

2) GPS tracking
Global Positioning System (GPS) units in smartphone
communicate with the satellites to work out our precise location
Fig. 1. Existing fall detection system
on Earth. The GPS technology doesn’t really use net knowledge
this can be why we are able to realize our location on maps even
4. Proposed system once losing the signals, however the map itself is foggy because
This study proposes a low-priced fall detection system, it needs net to load the small print. this can be however offline
victimisation the present devices and wireless technology, map works. GPS is employed altogether location-based apps
while not the necessity for hardware modification, like Uber and Google Maps.
environmental setup, and sporting external sensors. The
projected system has 3 central components: sensing, learning,
and alerting. within the initial element, we tend to take the
benefits of the knowledgebase-based approach to gather
realistic fall data. That is, we tend to collect the important
measuring system knowledge from the mobile embedded
sensors and record the corresponding user behavior to see the
desired parameters. within the second element, the projected
system learns the link between the autumn behavior and
therefore the collected knowledge. during this step, we tend to
utilize totally different fall options, together with vertical and
total acceleration, to style totally different fall detection
algorithms. we tend to additionally live the performance in each
sensitivity and specificity whereas considering their trade- off.
within the third element, the mobile phones alert pre- organized
emergency contacts through message. we tend to additional Fig. 3. GPS tracking
 International Journal of Research in Engineering, Science and Management 24
Volume-3, Issue-4, April-2020
www.ijresm.com | ISSN (Online): 2581-5792

3) Alert message emergency. The topmost number that is entered is called as

SMS messages are ancient text messages that support a wider Master contact. These emergency contact can be updated
vary of devices, however are restricted to solely short plain text anytime as per user’s wish. During an emergency situation, the
notifications. where your workers have a transportable signal person has two ways to indicate it. She has to either shake to the
they'll receive one in every of these alerts. SOS Alerts aim to mobile phone with force or long press the lower volume button
form emergency info a lot of accessible throughout a natural or continuously. Once the accelerometer /sensor detects this
human-caused crisis. we tend to gather relevant and vibration, the application immediately picks the current
authoritative content from the net, social media, and Google location and simultaneously picks the emergency contacts from
merchandise, and so highlight that info on Google merchandise the database and sends a SMS with a help message and G-map
like Search and Maps. looking on the character of the crisis and link of the fetched location to the emergency contacts. In
your location, you’ll see updates from native, national, or addition to this a call will be made to the master contact.
international authorities. These updates might embrace
emergency phone numbers and websites, maps, translations of
helpful phrases, donation opportunities, and more.

. Fig. 6. Architecture diagram

Fig. 4. Identifying exact location

Fig. 7. Flow diagram

5. Modules
A. Module 1: Authentication module
A registration page can open wherever we will enter the user
details. This contains details like Name of the user, Address of
the user, his/her Mobile range and Email-id. These details area
Fig. 5. Alert message sent through SMS
unit saved by clicking on save button. it's been designed by
keeping all the constraints in mind like, the mobile range is
The basic skeleton working of the application is shown in the
restricted to ten digits and user has to enter solely the digits,
Fig. 6.
email-id is ready victimization commonplace email-id format,
Initially user has to register in the application by providing
if user enters some wrong format or misses @ or .com then it
their basic detail such as name, email id, mobile number.
flashes a slip-up.
Permission to access contacts, messaging and GPS must be
enabled. Once registration is done the user can login with the B. Module 2: Adding emergency contact
valid credentials. Now moving on to the foremost essential part After registering user info, User will add the contact details
of the application, where the user has to enter the contact of his/her shut associates like members of the family or friends,
numbers of the trusted people who should be contacted during United Nations agency will reach instantly for facilitate just in
 International Journal of Research in Engineering, Science and Management 25
Volume-3, Issue-4, April-2020
www.ijresm.com | ISSN (Online): 2581-5792

case of emergency. A user will add new contacts with associate E. Module 5: Send SMS
degree emergency message that contains Mobile range. which By pressing volume key, Women security app will send the
might be viewed shortly when clicking on update contact button emergency message with full address to all the contacts added
of the applying. These 5 registered numbers can receive the in the emergency contact list. If both the internet and location
message. of victim’s android phone will have turned on then GPS will
fetch the location automatically and send an emergency
message to all contacts with the exact location of victim and
their landmark with longitude and latitude through messages.
So that the nearby associate can reach the victim easily for Her
help.

Fig. 12. Send SMS

Fig. 9. Emergency contact
6. Future enhancement
C. Module 3: Pressing volume button/shaking
Now a day’s women are facing many problems like rapes,
The last and most significant feature of this application is
robbery and sexual assault. These victims of violence have been
causation facilitate message and creating decision by merely
increased rapidly day by day in our country. Thus women
pressing volume key. when pressing volume key, a tiny low
hesitate to come out of home. They are mentally becoming
screen can seem for show activate GPS or not. when continuing,
weak to handle any such situations and populations of women’s
all the quantity in contact list of lady security app can solely
have been decreased. Hence, our secure app will help them to
receive the emergency message with full address. A decision is
overcome the above problems. The proposed app is tested in
created to the primary master contact.
different location and obtained a satisfactory result. In future
work, we are trying to enhance our app by using the send the
SMS without network.

7. Conclusion
In this paper, a low-cost women tracking system exploitation
GPS and GPRS of GSM network, appropriate for wide
selection of applications everywhere the globe. the mix of the
GPS and GPRS provides continuous and real time trailing. the
Fig. 10. Shake value is way lower compared to SMS primarily based trailing
D. Module 4: GPS tracking systems.
If the location of smart phone is turned on but if there is no References
internet connection at the time of pressing volume key, the
[1] Ahsan Shahzad and Kiseon Kim, “FallDroid: An Automated Smart-
emergency message will be send with longitude and latitude of Phone-Based Fall Detection System Using Multiple Kernel Learning”
victim’s sim's last network fetched by GPS. IEEE Trans. Ind. Informat., vol. 15, no. 1, pp. 35 - 44, Jan. 2019.
[2] C. Wang et al., “Low-power fall detector using triaxial accelerometry and
barometric pressure sensing,” IEEE Trans. Ind. Informat., vol. 12, no. 6,
pp. 2302–2311, Dec. 2016.
[3] J. Dai, X. Bai, Z. Yang, Z. Shen, and D. Xuan, “Perfalld: A pervasive fall
detection system using mobile phones,” in Proc. 8th IEEE Int. Conf.
Pervasive Comput. Commun. Workshops., 2010, pp. 292–297.
[4] F. Sposaro and G. Tyson, “ifall: An android application for fall
monitoring and response,” in Proc. Annu. Int. Conf. IEEE Eng. Med. Biol.
Soc., 2009, pp. 6119–6122.
[5] Mihail Popescu, Yun Li, Marjorie Skubic, and Marilyn Rantz, “An
acoustic fall detector system that uses sound height information to reduce
Fig. 11. GPS tracking
 International Journal of Research in Engineering, Science and Management 26
Volume-3, Issue-4, April-2020
www.ijresm.com | ISSN (Online): 2581-5792
the false alarm rate,” 30th Annual International IEEE EMBS Conference, IEEE Engineering in Medicine and Biology 27th Annual Conference,
August 2008. pages 3551–3554, September 2005.
[6] Shaou-Gang Miaou, Pei-Hsu Sung, and Chia-Yuan Huang, “A [8] N. Noury, P. Barralon, G. Virone, P. Boissy, M. Hamel, and P. Rumeau,
customized human fall detection system using omni-camera images and “A smart sensor based on rules and its evaluation in daily routines,” in
personal information,” Proceedings of the 1st Distributed Diagnosis and Proc. 25th Annu. Int. Conf. IEEE Eng. Med. Biol. Soc., 2003, vol. 4, pp.
Home Healthcare (D2H2) Conference, pp. 39–41. April 2006. 3286–3289.
[7] Jay Chen, Karric Kwong, Dennis Chang, Jerry Luk, and Ruzena Bajcsy,
“Wearable sensors for reliable fall detection,” Proceedings of the 2005