Smart Street System

A PROJECT REPORT ON
SMART STREET SYSTEM
in partial fulﬁllment of the requirement for the degree of
BACHELOR OF TECHNOLOGY
in
Computer Science & Engineering Of
Mahatma Ghandhi University
By
Libin Thomas(13012892)
Team Members
Kuriakose Mathew(13012891)
Midhun Shaji(13012897)
Sharun Santhosh(13012928)
Under the guidance of
Fr. Dr. Jaison Paul Mulerickal CMI
Mr. Joseph John
DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING
RAJAGIRI SCHOOL OF ENGINEERING & TECHNOLOGY
RAJAGIRI VALLEY, KOCHI 682-039

RAJAGIRI SCHOOL OF ENGINEERING & TECHNOLOGY
RAJAGIRI VALLEY, KOCHI 682-039
DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING
Certificate
Certified that this is a Bonafide Record of the Main project on Smart
Street System done by Libin Thomas, University register number 13012892
of branch Computer Science & Engineering during the Semester Eight
Year 2017 at Rajagiri School of Engineering and Technology, Kakkanad,
Kochi.
Project Guide Project Coordinator Head of Department

Smart Street System Project Report
ACKNOWLEDGMENT
We are immensely grateful to the Almighty God for His blessings and for helping us complete this
project successfully.
We express our heartfelt gratitude to Dr. A.Unnikrishnan,Principal,Rajagiri School of Engineering
and Technology and to Ms. Shimmi Asokan,Head of the Department,Computer Science & Engineer-
ing, RSET for providing us with the means and facilities for the completion of the project. We would
like to express Our sincere gratitude to our guide Fr Jaison Paul Mulerickal CMI,Professor,Department
of Computer Science & Engineering,RSET and Mr.Joseph John,Assistant Professor,Department of
Computer Science & Engineering,RSET for their valuable guidance and encouragement in pursuing
this Project. We sincerely thank all other technical teaching and non-teaching staff who extended
their help at all points throughout. Last but not the least, we take this opportunity to put across our
deepest gratefulness to our friends and to our parents, whose encouragement and support helped us a
lot for the successful completion of this project.
Department of CSE,RSET i

ABSTRACT
Smart street system is a IOT based system for automation of street light system and real time
monitoring of streets. All street light systems are interconnected with each other and also to a server
using IoT. Aim of the project is to develop a system that makes our streets smart. Developing a street
light system that is efficient and reliable. Efficiency is attained by developing systems that aim at
energy conservation. Conventional source of energy is wasted with inefficient system that remains
ON during day and not working at night, Automatic ON/OFF system is a solution for the same. Status
of each street light can be accessed from anywhere and anytime through Internet. Brightness control
part save energy by using the precise amount of light needed. System ensures error reporting and
maintenance. Maintenance costs can be cut short with real time fault monitoring and by using detailed
operational intelligence to improve day-to-day effectiveness and planning. Real time monitoring of
streets helps to develop a security system that is aimed at preventing antisocial elements. Video
captured by cameras installed at different locations in street is monitored for suspicious activities.
Department of CSE,RSET ii

Contents
1 INTRODUCTION 1
1.1 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Project Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.3 Design and Implementation Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.4 Assumptions and Dependencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.5 Development Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2 Literature Survey 5
3 System Speciﬁcation 7
3.1 System Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.2 System Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.2.1 Sensors and Camera . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2.2 Microcomputer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.2.3 Processing of data in cloud . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.2.4 Reporting errors and unusual activities . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.2.5 Controlling the lamps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.3 System Functions or Use Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.3.1 Use case:Monitor System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.3.2 Use case:Usage analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.3.3 Use case:Check for unusual activities . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.3.4 Use case:Report unusual activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.4 Non-functional Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.5 Flow diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4 Project Schedule 17
4.1 Work Breakdown and responsibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.2 Schedule, Milestones and Deliverables . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5 Implementation 18
5.1 Controlling of street lamps with the received sensor values. . . . . . . . . . . . . . . . . 18
5.1.1 System for controlling ON OFF of lamps . . . . . . . . . . . . . . . . . . . . . . . . . 18
5.1.2 Brightness control on sensing trafﬁc . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

5.1.3 Error reporting system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
5.2 Pushing values to cloud . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
5.2.1 Relayr cloud . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
5.2.2 Controlling the Second Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.3 Video analytics & Surveillance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.3.1 Automated Surveillance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.3.2 Opencv-Python . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.3.3 Motion detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
5.3.4 Masking Frames . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.3.5 Live Streaming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.3.6 Website . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6 Testing and Test Results 30
6.1 Testing strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
6.2 Summary of results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
7 Conclusion 38
8 Future Enhancements 39
9 Glossary 40
Bibliography 44
Annexure 45
Department of CSE,RSET iv

List of Figures
3.1 System architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.2 Sunlight intensity chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.3 GPIO pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.4 use case . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.5 Monitor system GUI(wireframe) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.6 Check for unusual activity GUI(wireframe) . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.7 Flow Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4.1 Work schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5.1 Relayr developer Dashboard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
5.2 Relayr Cloud Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
9.1 LDR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
9.2 PIR sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
9.3 Raspberry pi board . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

Chapter 1
INTRODUCTION
1.1 Problem Statement
To make our streets smart by bringing in energy efficient street light automation system and Real
time monitoring of streets to detect and prevent antisocial elements.
1.2 Project Scope
Street light system in our country is inefficient to large extent. This is due to either lack of tech-
nologies, unawareness and negligence. Smart street light system tries to find solution for the faster
depletion of energy resources due to the inefficient usage and negligence. Street light that remain on
during day time and lights that works at full brightness even when there is no traffic is something
that is related to wastage of energy. Less maintenance of this street light system is another problem
that arises due to unreported errors. A single system that receives the sensor value is send to other
similar systems hence reducing the number of sensors required. Anti-social elements are becoming
a burden for our nation where public infrastructure development is in a slow pace. Destroying public
properties and crime against women are of concern. Phase 2 helps with this by detecting activities
that are not normal in a street.
1.3 Design and Implementation Constraints
There are various challenges involved in this project.
• Maximum distance to which the sensor can sense is limited to 1 Metre, which may not be sufficient
for a 45 metre road.
• Camera modules are costly and each video received has to be processed.
Department of CSE,RSET 1

• Real time video monitoring and processing requires large storage.
• Detection of unusual activity is possible with only some incidents such as climbing a wall or an
accident.
• Accuracy of the sensors used are limited, especially of the proximity sensors. Higher accuracy
senors can be used but this may increase the overall cost.
• Large scale implementation would require to handle lamps that works at higher voltages and con-
trolling such lamps can be done with additional circuitrys, this then account to increase in cost.
1.4 Assumptions and Dependencies
• WIFI connection should be available continously with less interruption.
• Uninterrupted power supply for Raspberry pi.
• Processing in cloud is ﬂexible.
• External interference to the hardware components is not present.
• Larger storage available in cloud.
• Pushing the information to the cloud can be done at any point of time.
• All the processes are expected to be synchronous and real time.
1.5 Development Method

A software development process, also known as a software development life cycle (SDLC), is a
structure imposed on the development of a software product. Similar terms include software life cy-
cle and software process. It is often considered a subset of systems development life cycle. There are
several models for such processes, each describing approaches to a variety of tasks or activities that
take place during the process. Some people consider a life-cycle model a more general term and a
software development process a more specific term. For example, there are many specific software
development processes that ’fit’ the spiral life-cycle model. ISO/IEC 12207 is an international stan-
dard for software life-cycle processes. It aims to be the standard that defines all the tasks required for
developing and maintaining software. There are many different models and methodologies, but each
generally consists of a series of defined steps or stages.
Few popular models are:
• Waterfall Model
• V-Model
• Spiral Model
• Prototyping
• Iterative (Incremental Delivery)
• Rapid Application Development
From these we choose to follow Incremental model. In incremental model the whole requirement
is divided into various builds. Multiple development cycles take place here, making the life cycle a
"multi-waterfall" life cycle. Cycles are divided up into smaller, more easily managed modules. Each
module passes through the requirements, design, implementation and testing phases. A working
version of software is produced during the first module, so you have working software early on
during the software life cycle. Each subsequent release of the module adds function to the previous
release. The process continues till the complete system is achieved.
The incremental build model is a method of software development where the model is designed,
implemented and tested incrementally (a little more is added each time) until the product is finished.
It involves both development and maintenance. The product is defined as finished when it satisfies
all of its requirements. This model combines the elements of the waterfall model with the iterative
philosophy of prototyping. The product is decomposed into a number of components, each of which

are designed and built separately (termed as builds). Each component is delivered to the client when
it is complete. This allows partial utilization of product and avoids a long development time. It also
creates a large initial capital outlay with the subsequent long wait avoided. This model of develop-
ment also helps ease the traumatic effect of introducing completely new system all at once.
• Advantages of Incremental model:
> Generates working software quickly and early during the software life cycle.
> This model is more ﬂexible â ˘A¸S less costly to change scope and requirements.
> It is easier to test and debug during a smaller iteration.
> In this model customer can respond to each built.
> Lowers initial delivery cost.
> Easier to manage risk because risky pieces are identiﬁed and handled during itâ ˘A´Zs itera-
tion.

Chapter 2
Literature Survey
Phase 1 implements a single system that does three functions mainly automatic switching, bright-
ness control and error reporting. There are existing system that does any one of the function like
The LACS system contains a centralized or distributed architecture determined by application re-
quirements and space usage. The system optimizes the calculations and communications for light-
ing intensity, incorporates user illumination requirements according to their activities and performs
adjustments based on external lighting effects in external sensor and external sensor-less architec-
tures[1].The street lighting system is one of the responsible for overloading the power system during
the peak load time, because the cityâ ˘A´Zs lights are turned on around this time[3]. In order to mon-
itoring and control street lighting, the lighting pole controller was designed and implemented based
on wireless sensor network. The controllers were installed at each lighting pole.The hardware of
controller integrates a AVR microcontroller, a radio modem, double lamp control units and current
detecting units. The lamp control units control primary and secondary lamp on or off according to re-
ceiving command from RTU. The current detecting units measure the power line current to determine
the lamp status which reporting to the center[4].The growing interest towards green and efficient use
of electrical energy has recently pushed the industry of street light control systems. In the past, very
simple on/off switching mechanism based on daylight sensing and cyclic preventive maintenance
procedures were adopted. Nowadays, intelligent control systems offering remote supervision have
strongly contributed to a change in perspective in maintenance engineering form a traditional "Fail
and Fix" view to a "Predict and Prevent" approach[5].Hence techniques to control it has to be devised.
Ausgrid, a state owned electricity infrastructure company in Australia and New South Wales[6]. It
is possible to report errors in Ausgrid website. But here the reporting is not done by the system, a
person who sees the same has to report the error. This is not effective as there may be street light
systems in rural areas where people find it difficult accessing the web and remain unreported.
Another existing system is the intelligent street light system in Oslo referred to as Oslo project.
Oslo project have street lights that dim on detection of movement. This project has gone further
by enabling street lights to communicate with each other. But the error reporting system is still not
included. A survey of the existing systems revealed that there is inability in implementing all three

functions in a street light system.
Specialist video software, in use at the Farsight Observatory, now automatically and autonomously
detects unusual activity[7]. Without any manual input, analytical video can detect activity such as: an
individual lingering at an unmanned petrol station someone suspiciously inspecting your perimeter
security an unknown vehicle parked in a controlled zone . Security operators at the Farsight Ob-
servatory will be able to rapidly respond to alarms, already know their cause, whilst CCTV cameras
continue to track and monitor movement. For example, after an alarm has been raised CCTV cameras
will pan, tilt and zoom to track an intruder moving across private property. In this system the security
is ensured but not extended to the level where the concerned authority receives a notification.
Road traffic video analytics aims at using a number of techniques to achieve better traffic and
road safety, control congestion and provide immediate care for accident victims[5]. In this paper,a
near real-time traffic analytics system which can automatically detect road accidents from live video
streams is proposed. The system alerts nearby hospitals and highway rescue teams when accidents
occur. It also detects road congestion and broadcasts alternative route information to relevant com-
muters.Apache hadoop was used to implement the above system. The Apache Hadoop software
library is a framework that allows for the distributed processing of large data sets across clusters of
computers using simple programming models[8]. It is designed to scale up from single servers to
thousands of machines, each offering local computation and storage. Rather than rely on hardware to
deliver high-availability, the library itself is designed to detect and handle failures at the application
layer, so delivering a highly-available service on top of a cluster of computers, each of which may be
prone to failures.
Existing CCTV systems are not complete and have not evolved with the emerging technologies.
Real time monitoring of streets is important when anti-social elements like â ˘AŸRail hoonsâ ˘A´Z, a
gang of graffiti artists defacing rail properties, are causing problems. Phase 2 of the project can be
developed as an addition to the existing security systems. One of the major problem with video
surveillance is that it has a chilling effect on public life, this can be solved by having surveillance
system where no one but a computerized system is looking at you when you are in a street. Hence
encouraging legitimate exercise of natural and legal rights. Video surveillance have many drawbacks
that has to be dealt with for effective implementation of the system. Smart street system is a model
for the real world implementation which if implemented can be a good change in the existing street
light and street security systems.

Chapter 3
System Speciﬁcation
3.1 System Overview
To make the smart street light system work the data from ldr and pir sensors are taken and the data
from the individual nodes are pushed onto a central cloud where the data is assimilated and processed
further and mapped on to a common maps platform. Ldr will take readings of the light intensity,PIR
sensors will work inorder to detect any movement.
To detect any unusual behaviour in the streets the real time videos are processed over the apache
Hadoop platform with the help of MapReduce framework and errors get reported over the cloud.
3.2 System Architecture
Figure 3.1: System architecture

System is divided into the following:
• Sensors and Camera.
• Microcomputer(Raspberry pi).
• Processing of data in cloud.
• Reporting Errors and unusual activities.
3.2.1 Sensors and Camera
Sensors used in the system include LDR and PIR. LDR sensors are used to check the intensity of
lights and accordingly control the lamp in ON/OFF state.When the intensity of the light falling on
the sensor is less than the preset threshold value then the lamps are turned ON and if it is greater than
the preset value then the lamp is in OFF state.

Figure 3.2: Sunlight intensity chart
LDR sensor along with a 1K resistor gives an analog input value to the input pin this input value
has to be converted to digital value and has to be compared with the threshold value.LDR sensors are
also used to detect whether the lamp is working or not. this can be done by comparing the output
value from the sensor with the preset threshold value. If the input from the ldr facing the lamp is
below the preset value even after the instruction to turn ON the lamp has been send then it can be
detected as an error.This error can then be reported to the admin.PIR sensors are used to detect any
movement. PIR sensors are placed facing the road or the walkway. If the PIR sense movement then
it will have a high value and if not a low value. Once the Lamp is in on state then the input from the
PIR is checked if the value is not high for a particular time period then the lamp is brought to low
brightness state. Once the PIR is high then the lamp returns to high brightness state. The time period
for which the lamp is in low brightness state can be set to 15 minutes that is once the lamp is in on
state it continues in full brightness for the next 15 minutes after which the PIR value is checked and
if it is a low value then the brightness is reduced. there are different camera modules available that
are compatible with Raspberry pi boards. camera module have a cable that slots into the connector
situated between the Ethernet and HDMI ports, with the silver connectors facing the HDMI port. It
is possible to record videos using start_recording() and stop_recording() functions after conﬁguring
the camera module.

3.2.2 Microcomputer
Microcomputer used is a Raspberry Pi 2 board which is credit card sized microcomputer that can be
used for different processing and computations. it has a row of GPIO (general purpose input/output)
pins.These pins are a physical interface between the Pi and the outside world. At the simplest level,
you can think of them as switches that you can turn on or off (input) or that the Pi can turn on or
off (output). 26 of the 40 pins are GPIO pins; the others are power or ground pins. Output from
the sensors can be fed to these GPIO pins and then can be used for different functionality including
pushing to cloud.
Figure 3.3: GPIO pins
3.2.3 Processing of data in cloud
Data that is pushed to cloud include the sensor values. Sensor values are used as input to functions
to control the lamp. Results from the processing are send to the next modules. Relayr cloud platform
was used to receive sensor data. Raspberry system is add Cloud computing as a distributed computing
paradigm, provides an environment to perform large-scale data processing.

3.2.4 Reporting errors and unusual activities
The errors if any are reported to the admin,which include the error of the lamp. If senor values are
not received it can be considered as a system error. Unusual activities detected has to be reported to
the admin and this has to be a quick action which remains a challenge in the entire functioning of the
project. Error reporting use case has been detailed in the next section.
3.2.5 Controlling the lamps
The resistance of an LDR is very high, sometimes as high as 1000 000 ohms, but when they are
illuminated with light resistance drops dramatically. When the light level is low the resistance of the
LDR is high. Sunlight intensity less than 250 lux. Sunlight intensity greater than 250 lux. lamps are
then turned ON/OFF accordingly PIR sensor detects no movement. Once no movement is detected
while the light remain on for more than 15minute then the brightness is reduced using the brightness
reducing circuitry If there is any movement detected by the PIR sensor then increase the brightness
by controlling voltage Continue this for next 15 minutes then resume the previous state Once the ldr
in front of the lamp detects tha light is not on after the instruction for the same has been send this can
be an error

3.3 System Functions or Use Cases
Figure 3.4: use case
The functional requirements for the user and admin to interact with the open platform or the web-
site are mapped as the use cases shown in the Figure 2.1. Admin monitor the entire system for proper
functioning. Errors can occur when a street light is not functional, in which case the microcomputer
sets message for admin and push to cloud. Error can also occur when the microcomputer is not func-
tioning this can be detected when there is no value being pushed to cloud. Errors with street light are
reported to the electricity section for maintenance while the error with the microcomputer has to be
taken care of by the system maintenance wing. Once an unusual activity detected the the notiﬁcation

pops up followed by reporting to concerned authorities.
3.3.1 Use case:Monitor System
The admin can monitor the system to check the status of all the street lamps in a locality.
Figure 3.5: Monitor system GUI(wireframe)
Pre-Conditions: System works without any errors.
Post-Conditions: Error in the street light or the microcomputer detected and reported.
3.3.2 Use case:Usage analysis
Analysis include the unit of electricity used by the system this can be done by tracking the time for
which the street lights works and the time for which it is in dim state and full brightness state.

Pre-Conditions: sensor values pushed to cloud continuously.
Post-Conditions: Energy usage analysed saved.
3.3.3 Use case:Check for unusual activities
admin receives a notification for unusual activity. Record of such activities is maintained for fur-
ther use.
Figure 3.6: Check for unusual activity GUI(wireframe)
Pre-Conditions: Camera module working and no unusual activity.
Post-Conditions: Recording the details of the unusual activity.
3.3.4 Use case:Report unusual activities
System reports any unusual activity to the concerned authority and the admin receives a notifica-
tion. Admin then make sure that the concerned authority has received the message. And give public
notification if necessary.

Pre-Conditions: Details of the incident available.
Post-Conditions: Concerned authority notified.
3.4 Non-functional Requirements
• When the admin uses the website, he/she should be able to interact smoothly with the site.
• There should be a working internet connection for the cloud pushing, this can be made available
through a wired connectivity.
• Admin should receive notification of unusual activity quickly. This also should ensure that the
concerned authority receives notifications quickly.
• Communication of admin with police department and electricity section should be smooth for
quick actions.
• The internet connection should be fast enough to perform video analytics and get reports real time.
• For the above implementations the cloud should be reliable and flexible for cloud pushing and
processing.
• Access for users, other than admin, to the website for live monitoring is outside the scope of the
present phase of the project.

3.5 Flow diagram
Figure 3.7: Flow Diagram

Chapter 4
Project Schedule
4.1 Work Breakdown and responsibility
Modules included in the project are:
• Sensor setup, analysis of the values received and setting up the threshold value for detection.
Installing and testing of camera.
•Pushing the received values to cloud and analysis of the data in the cloud platform.
• Front end development for receiving notiﬁcations.
• Video analytics which include detection and reporting unusual activity from the video input.
First two are implemented in phase 1 and the next two as phase 1. Each member is assigned with a
module and divide the work equally.
4.2 Schedule, Milestones and Deliverables
Figure 4.1 show the project schedule.
Figure 4.1: Work schedule

Chapter 5
Implementation
Functionalities and effectiveness of the proposed system were tested under various lighting con-
ditions as well as for various situation of motion detection. The proposed lighting system worked
efﬁciently under all the given conditions and was found be more effective than the present street light
system. The various modules implemented in this project work are:
• Controlling of street lamps with the received sensor values.
• Pushing values to cloud.
• Video analytics and reporting.
5.1 Controlling of street lamps with the received sensor values.
Sensors used in the system are LDR(Light Dependent Resistors) and PIR(Proximity InfraRed).When
the intensity of the light falling on the sensor is less than the preset threshold value then the lamps are
turned ON and if it is greater than the preset value then the lamp is in OFF state.LDR sensors are also
used to detect whether the lamp is working or not, this can be done by comparing the output value
from the sensor with the preset threshold value.If the lamp is not working, then a e-mail alert is sent
to Admin.
5.1.1 System for controlling ON OFF of lamps
The resistance of an LDR is very high, sometimes as high as 1000 000 ohms, but when they are
illuminated with light resistance drops dramatically. When the light level is low the resistance of the
LDR is high.
Pseudocode:
pinMode(A0,INPUT);
void setup() {

pinMode(4,OUTPUT);
pinMode(A1,INPUT);
}
void loop() {
if(analogRead(A0)<400){
digitalWrite(4,HIGH);
if(analogRead(A1)<20) {
SendMessage();
}
}
else {
digitalWrite(4,LOW);
}
}
5.1.2 Brightness control on sensing trafﬁc
PIR sensor detects no movement. Once no movement is detected while the light remain on for
more than 15minute then the brightness is reduced. Switching of LED between the bright and the
dim mode is achieved by PWM output from raspberry pi.PWM â ˘A¸SPulse Width Modulation or Pulse
Duration Modulation.It is a modulation technique used to encode a message into a pulsing signal.
Its main use is to control the power supplied to the devices. PWM uses duty cycle to control the
power.A low duty cycle corresponds to low power and a high duty cycle corresponds to high power.It
is normally represented in percent, 100% being fully on. Main advantage of PWM is that power loss
in switching devices is very low. If there is any movement detected by the PIR sensor then increase
the brightness by controlling voltage. Continue this for next 15 minutes then resume the previous
state Code for sending message through gsm:
Pseudocode:
if(ldr2value<ld2th)
{
if(irvalue<500 && ldr1value<ldrth)
{
dim(dutycycle);
delay(.1);

}
else
digitalWrite(6,HIGH);
irvalue=analogRead(A2);
ldr1value=analogRead(A0);
}
5.1.3 Error reporting system
Once the ldr in front of the lamp detects that light is not on after the instruction for the same has
been send then a mail is send using smtp protocol.
Email alert pseudocode:
import smtplib
from email.mime.text import MIMEText
server = smtplib.SMTP(’smtpmail.provider.com’, 587)
server.ehlo()
server.starttls()
server.ehlo
server.login("email or username", "Password")
value = "Critical warning!"
msg = MIMEText(value)
msg[’Subject’] = "Critical warning!"
msg[’From’] = "Raspberry Pi"
msg[’To’] = "recipient@gmail.com"
server.sendmail("consigner@gmail.com", "recipient@gmail.com", msg.as_string())
server.quit()
5.2 Pushing values to cloud
Smart street system use Relayr Cloud Platform.The relayr Cloud facilitates all of the interactions
between connected devices and the internet.Relayr is used to establish Communication between two
raspberry pi boards.The LDR readings of ﬁrst Pi will be pushed to cloud and will be received by the

second Pi.
5.2.1 Relayr cloud
The relayr Cloud is the core of the relayr IoT middleware platform. Starting point for using the relayr
Cloud is the Developer Dashboard: the user interface of the Cloud. Once registered, it provides
with an interface for managing cloud-connected devices, watching the data coming in, and sending
commands to those devices.
Figure 5.1: Relayr developer Dashboard
The diagram below illustrates the architecture of the relayr Cloud and the ﬂow of data to and from
users and devices

Figure 5.2: Relayr Cloud Architecture
Devices transmit and receive data to and from the Cloud through one or more application pro-
tocols (e.g., MQTT and Websockets). The data is passed through the MQTT Broker, which uses the
Data Router to ensure that the data is delivered to the proper recipients, such as the data store and
services that subscribe to that deviceâ ˘A´Zs data. Cloud users can access the Cloud directly through
the HTTP API or through the Developer Dashboard, which provides an interface for accessing device
data and services.
The following entities reside outside of the Cloud and use one or more application protocols to
connect and interface with it:
• User:
A user is an individual or an external application who is using the relayr platform.
• Devices:
Devices constitute the most basic element of the relayr platform. A device is an entity that can
transmit and receive data to and from the Cloud. As shown in the diagram, devices can connect
to the Cloud through MQTT and Websockets. However, in a fog architecture setup, devices
can also connect through Vertex.
• Developer Dashboard:
The Developer Dashboard is the user interface of the relayr Cloud. It provides a visual interface
for managing devices and accessing device data.

Application protocols facilitate the connection between clients (users, devices, the Dash-
board) and the Cloud.
• HTTP API:
relayr´s HTTP API provides programmatic access to all of the resources and services in the
relayr Cloud. When a user or application makes a call to the API to execute a command or
return a resource, the call is routed to the correct endpoint or service through the HTTP router,
and then the desired action is performed.
• MQTT:
MQTT is an industry-standard lightweight application protocol that uses a pub/sub messaging
pattern. It is currently the primary application protocol used by the relayr Cloud for device data
transmission.
Generated Firmware:
import time
import paho.mqtt.client as mqtt
# mqtt credentials
creds = {
’clientId’: ’T1cdngsseRyaxnACQ+3wRJA’,
’user’: ’d5c76782-cb1e-4726-b19c-0090fb7c1124’,
’password’: ’G.MCbXT0CjgE’,
’topic’: ’/v1/d5c76782-cb1e-4726-b19c-0090fb7c1124/’,
’server’: ’mqtt.relayr.io’,
’port’: 1883
}
publishing_period = 1000
class MqttDelegate(object):
#A delegate class providing callbacks for an MQTT client.
* def __init__(self, client, credentials):
self.client = client
self.credentials = credentials
def on_connect(self, client, userdata, ﬂags, rc):
print(’Connected.’)
self.client.subscribe(self.credentials[’topic’] + ’cmd’)

def on_message(self, client, userdata, msg):
print(’Command received: %s’ % msg.payload)
def on_publish(self, client, userdata, mid):
print(’Message published.’)
def main(credentials, publishing_period):
client = mqtt.Client(client_id=credentials[’clientId’])
delegate = MqttDelegate(client, creds)
client.on_connect = delegate.on_connect
client.on_message = delegate.on_message
client.on_publish = delegate.on_publish
user, password = credentials[’user’], credentials[’password’] client.username_pw_set(user,
password)
try:
print(’Connecting to mqtt server.’)
server, port = credentials[’server’], credentials[’port’]
client.connect(server, port=port, keepalive=60)
except:
print(’Connection failed, check your credentials!’)
return
# set 200 ms as minimum publishing period if publishing_period < 200:
publishing_period = 200
while True:
client.loop()
# publish data
message = {
’meaning’: ’py temperature’,
’value’: ’py something’
} client.publish(credentials[’topic’] +’data’,
json.dumps(message))
time.sleep(publishing_period / 1000.)
if __name__ == ’__main__’:
main(creds, publishing_period)
Raspberry pi is not available as a device in Relayr platform.Solution is to choose another device

and edit it to suit pi.The data is pushed to the cloud using the MQTT protoco l.MQTT is a
machine-to-machine (M2M)/"Internet of Things" connect ivity protocol. It was designed
as an extremely lightweight publish /subscribe messaging transport.Once the device is
added to the relayr platfor m and executing the code for pushing the data, the relayr Cloud
will recognize the data coming in from the virtual device and display it o n the Dashboard.
5.2.2 Controlling the Second Device
The second Pi can be controlled by deﬁning a rule inside relayr.The data can also be retrieved by
using curl command.
curl -X GET -H "Authorization: Bearer H.YR0gcgNvy-21kc2AInfHngnB.tzXwh" -H "Cache-
Control: no-cache" "https://api.relayr.io/devices/c298b922-148c-45d0-ad14-7afc9620bc42/readings"
5.3 Video analytics & Surveillance
This module is designed to work in two modes.
• Automated Surveillance
• Live Streaming
5.3.1 Automated Surveillance
Uses Opencv-python to carry out the processing of the video captured.Restricted region is masked
by setting those region intensities to zero.Numpy package is used.If any trespassing is detected within
the restricted area, then a screenshot of the same will be send as a mail to the concerned authority.Uses
Dropbox API to store the screenshots.DropboxClient package is used.E-mail alert send using smtplib
package in python. Processing involved in automated surveillance is detailed in following subsec-
tions.
5.3.2 Opencv-Python
Uses Opencv-python to carry out the processing of the video captured.Restricted region is masked
by setting those region intensities to zero.Numpy package is used.If any trespassing is detected within
the restricted area, then a screenshot of the same will be send as a mail to the concerned authority.Uses
Dropbox API to store the screenshots.DropboxClient package is used.E-mail alert send using smtplib
package in python.

Python is a general purpose programming language started by Guido van Rossum, which became
very popular in short time mainly because of its simplicity and code readability. It enables the pro-
grammer to express his ideas in fewer lines of code without reducing any readability. Compared to
other languages like C/C++, Python is slower. But another important feature of Python is that it can be
easily extended with C/C++. This feature helps us to write computationally intensive codes in C/C++
and create a Python wrapper for it so that we can use these wrappers as Python modules. This gives
us two advantages: first, our code is as fast as original C/C++ code (since it is the actual C++ code
working in background) and second, it is very easy to code in Python. This is how OpenCV-Python
works, it is a Python wrapper around original C++ implementation. And the support of Numpy
makes the task more easier. Numpy is a highly optimized library for numerical operations. It gives a
MATLAB-style syntax. All the OpenCV array structures are converted to-and-from Numpy arrays.
So whatever operations you can do in Numpy, you can combine it with OpenCV, which increases
number of weapons in your arsenal. Besides that, several other libraries like SciPy, Matplotlib which
supports Numpy can be used with this. So OpenCV-Python is an appropriate tool for fast prototyping
of computer vision problems.
5.3.3 Motion detection
The background of our video stream is largely static and unchanging over consecutive frames of a
video. Therefore, if we can model the background, we monitor it for substantial changes. If there is
a substantial change, we can detect it - this change normally corresponds to motion on our video[10].
The first frame of our video file will contain no motion and just background â ˘AˇT therefore, we can
model the background of our video stream using only the first frame of the video. First frame gets
updated in fixed interval of time to take into account the situation where an object may be placed in
the location, like a waste bin. –min-area is the minimum size (in pixels) for a region of an image to
be considered actual â ˘AIJmotionâ ˘A˙I. This is set to not detect minor motion such as a leaf motion in
wind.
Algorithm for video analytics:
Step 1: Start
step 2: Import packages picamera, cv2, numpy
step 3: Initialize the camera and grab a reference to the raw camera capture
step 4: Allow the camera to warm up, then initialize the average frame, last uploaded timestamp, and
frame motion counter

step 5: Capture frames from the camera
step 6: Grab the raw NumPy array representing the image and initialize the timestamp and occupied
or unoccupied text
step 7: Resize the frame, convert it to grayscale, and blur it
step 8: If the average frame is None, initialize it
step 9: Accumulate the weighted average between the current Frame and previous frames, then com-
pute the difference between the current frame and running average
step 10: Threshold the delta image, dilate the thresholded image to fill in holes, then find contours on
thresholded image
step 11: Loop over the contours if the contour is too small, ignore it
step 12: Compute the bounding box for the contour, draw it on the frame and update the text
step 13: Print datetime.datetime.time(datetime.datetime.now())
step 14: Display the frame on screen
step 15: If text is occupied send alert and upload image to dropbox
step 16: continue step 9 to step 15 until camera stops
step 17: stop
5.3.4 Masking Frames
Mask operations on matrices are quite simple. The idea is that we recalculate each pixels value in an
image according to a mask matrix (also known as kernel). This mask holds values that will adjust
how much influence neighboring pixels (and the current pixel) have on the new pixel value. From a
mathematical point of view we make a weighted average, with our specified values.
mask = np.zeros(frame.shape,np.uint8)
mask[y:y+h,x:x+w] = frame[y:y+h,x:x+w]
Numpy is called as np in above code.Rectangle frame coordinates is marked using variables z,y,w
and h.
5.3.5 Live Streaming
This mode in video analytics and surveillance is concerned with live streaming of video captured
from the camera which is placed in the walkway or any street. Anyone in the world with an access
to the website will be able to watch what happens in a street from anywhere in the world. Access to
the website is given to only people who are supposed to do the surveillance such as someone in the
police department. Achievement in this area is that the video is accessible from anywhere and not

in a single system which is usually placed in some place near to the admin. Consider the situation
where police person receives notiﬁcation of some unusual activity and want t0 watch the street at that
moment. If some other system is used he has no option but to be at the place and clear the danger
and what if there was no danger at all and that was a wrong notiﬁcation. This is where Smart street
system is important with its live streaming of streets.
Feature of Dataplicity was used to achieve remote access to Raspberry pi shell.The Dataplicity
client uses a opportunistically-connected secure web connection to provide a communications chan-
nel between device and Dataplicity, and web browser attaches to the other end of that channel.It
works as follows.
Setting up Pi camera The Raspberry Pi camera can appear in /dev as a standard USB video device
using following command
sudo modprobe bcm2835-v4l2
Install required packages Install packages libjpeg8-dev imagemagick libv4l-dev
Use MJPG-streamer MJPG-streamer application is downloaded and is used fro live sreaming.
Run MJPG-streamer This can be done using the command
sudo ./mjpg_streamer -i "./input_uvc.so -f 10 -r 640x320 -n -y" -o "./output_http.so -w ./www
-p 80"
Activate wormhole in dataplicity Now the raspberry pi could be accessed through dataplicity.
Access camera video stream via Dataplicity Wormhole Accessed at the link
https://<DEVICE_ID>.dataplicity.io/stream_simple.html this link can be embedded in a script
to be accesed from the website designed for the surveillance.
5.3.6 Website
Website is designed and used to make the surveillance and automation system strong. Website
provides the folowing option regarding automation assuming that the admin has logged in using his
credentials and has access to the relayr dashboard.

Access to relayr dashboard This link in the Website redirects to the relayr dashboard. The LDR
and PIR values of each device can be monitored and analysed. It is possible to generate a graph of
values received. This can be later used to find patterns in lamp working in dim mode and indirectly
get the traffic in the street after a particlar time.
Check email refernce to the mail account of the admin. If admin uses a system in which he is the
root user and has logged in to the mail account, then admin is directed to inbox. Mail for Error and
unusual activity is received in the admin inbox. This then act as notification system.
Start live streaming The above address to remote raspberry access in dataplicity is embedded in
the script that is run when the option for starting live streaming is chosen. Multiple raspberry can be
connected and a list of available camera can be created as an enhancement of this module, as for now
there is only one camera connected.

Chapter 6
Testing and Test Results
6.1 Testing strategies
1. Automatic ON/OFF of lamp sensing sunlight intensity.
2. Reduced brightness of lamp on detecting no motion.
3. Lamp Error reporting through email.
4. Successful login to website
5. Detecting entrance to restricted area as an unusual activity.
6. Live streaming of street.
7. Pushing sensor Data to relayr cloud.
8. Email alert on unusual activity. 9. Image upload to Dropbox on unusual activity.

6.2 Summary of results
case 1:

case 2:

case 3:
case 4:

case 5:

case 6:

case 7:

case 8:
case 9:

Chapter 7
Conclusion
Energy efﬁcient technology and energy conserving systems is a need of the time when the energy
sources are getting depleted at faster rate. Smart street system is such a system that includes real time
monitoring of streets. Development of the project to something that can be applied in larger scale
will be a great achievement and enhance the existing technologies that aim at energy conservation.
Automated video surveillance and live streaming of streets can strengthen the surveillance system
and aims at preventing antisocial elements.

Chapter 8
Future Enhancements
Some of the features that can be added to the project in future include
• Website notification of unusual activity rather than email alert.
• Saving video for future reference.
Website notification Website notification was a feature proposed but was not able to implement due
to time constraints. This feature can be implemented such that admin receives a pop up indicating
the unusual activity along with the image of the incident. implemented email alert is sufficient but it
involves a mail service. website notification for error of lamps can also be implemented.
Saving video Live streaming is a strong way to implement surveillance from anywhere any time,
but it can be enhanced if the video could be saved to some database for future reference. This involves
large storage and may end up in system delay in capturing and sending.

Chapter 9
Glossary
Abbreviations
• LDR-Light Dependent Resistors
• PIR-Passive infrared sensor
• SDLC-Software Development Life Cycle
• GPIO-General purpose input output
• VA-Video Analytics.
Deﬁnitions
• LDR - All objects with a temperature above absolute zero emit heat energy in the form of radiation.
Usually this radiation isn’t visible to the human eye because it radiates at infrared wavelengths,
but it can be detected by electronic devices designed for such a purpose. Photo conductivity is
an optical phenomenon in which the materials conductivity is increased when light is absorbed
by the material. When light falls i.e. when the photons fall on the device, the electrons in
the valence band of the semiconductor material are excited to the conduction band. These
photons in the incident light should have energy greater than the band gap of the semiconductor
material to make the electrons jump from the valence band to the conduction band. Hence when
light having enough energy strikes on the device, more and more electrons are excited to the
conduction band which results in large number of charge carriers. The result of this process is
more and more current starts ﬂowing through the device when the circuit is closed and hence it
is said that the resistance of the device has been decreased. This is the most common working
principle of LDR[9]

Figure 9.1: LDR
• PIR-A passive infrared sensor is an electronic sensor that measures infrared (IR) light radiating
from objects in its ﬁeld of view. They are most often used in PIR-based motion detectors.All
objects with a temperature above absolute zero emit heat energy in the form of radiation. Usu-
ally this radiation isn’t visible to the human eye because it radiates at infrared wavelengths, but
it can be detected by electronic devices designed for such a purpose. The term passive in this
instance refers to the fact that PIR devices do not generate or radiate any energy for detection
purposes. They work entirely by detecting the energy given off by other objects.PIR sensors
don’t detect or measure "heat"; instead they detect the infrared radiation emitted or reﬂected
from an object.

Figure 9.2: PIR sensor
• Raspberry Pi - It is a small, bare-bones computer developed by The Raspberry Pi Foundation, a
UK charity, with the intention of providing low-cost computers and free software to students.
Their ultimate goal is to foster computer science education.
Figure 9.3: Raspberry pi board

• Video analytics-Video content analysis (also Video content analytics), is the capability of auto-
matically analyzing video to detect and determine temporal and spatial events. This techni-
cal capability is used in a wide range of domains including entertainment, health-care, retail,
automotive, transport, home automation, flame and smoke detection, safety and security.The
algorithms can be implemented as software on general purpose machines, or as hardware in
specialized video processing units. Many different functionalities can be implemented in VCA.
Video Motion Detection is one of the simpler forms where motion is detected with regard to a
fixed background scene. More advanced functionalities include video tracking and egomotion
estimation.
Based on the internal representation that VCA generates in the machine, it is possible to build
other functionalities, such as identification, behavior analysis or other forms of situation aware-
ness.
VCA relies on good input video, so it is often combined with video enhancement technologies
such as video denoising, image stabilization, unsharp masking and super-resolution.
• MJPG-Streamer- It is a tool for the command line to stream video data from a webcam or other
video source as a Motion JPEG (M-JPEG / MJPG). While modern network cameras automat-
ically generate such a stream, the program can also be used to convert a simple webcam using
a computer with Internet access to such a stream.

Bibliograghy
[1] Reza Mohamaddoust , Abolfazl Toroghi Haghighat, Mohamad Javad Motahari Sharif and Nic-
colo Capanni, â ˘AIJA Novel Design of an Automatic Lighting Control System for a Wireless
Sensor Network with Increased Sensor Lifetime and Reduced Sensor Numbersâ ˘A˙I, Sensors
(2011) ,Volume No.- 11(9), pp. 8933-8952.
[2] De Dominicis, C.M. Flammini, A.; Sisinni, E.; Fasanotti, L.; Floreani, F.; â ˘AIJOn the devel-
opment of a wireless self localizing streetlight monitoring system â ˘AIJ, Sensors Applications
Symposium IEEE, pp. 233 - 238 ,2011.
[3] Gustavo W. Denardin, Carlos H. Barriquello, Alexandre Campos, Rafael A. Pinto, â ˘AIJControl
Network for Modern Street Lighting Systemsâ ˘A˙I, IEEE symposium on Industrial Electronics
(ISIE), (2011), pp. 1282 â ˘A¸S 1289.
[4] Jing Chunguo, Wang Yan Sun, Wenyi Song,â ˘AIJDesign of Street Light Pole Controller Based
on WSNâ ˘A˙I, The Tenth International Conference on Electronic Measurement & Instruments,
ICEMI (2011), pp. 147 â ˘A¸S 150.
[5] Vaithilingam Anantha Natarajan,Subbaiyan Jothilakshmi,Venkat N Gudivada,"Scalable Trafﬁc
Video Analytics using Hadoop MapReduce"The First International Conference on Big Data,
Small Data, Linked Data and Open Data ALLDATA(2015)
[6] "https://www.farsight.co.uk/remote-monitoring-services/video-analytics", September 2016
[7] "https://www.ausgrid.com.au",March 2016
[8] "http://hadoop.apache.org/",March 2016
[9] "http://www.electrical4u.com/light-dependent-resistor-ldr-working-principle-of-ldr",November
2016
[10] "http://www.pyimagesearch.com",March 2017
[11] Adrian Rosebrock,"Image Search Engine Resource Guide"
[12] "https://www.dataplicity.com",March 2017

Remove the page

Smart Street System

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Smart Street System

Similar to Smart Street System (20)

Recently uploaded

Recently uploaded (20)

Smart Street System