Professional Documents
Culture Documents
Final Wo Format
Final Wo Format
on
1
ACADEMIC YEAR 2017-18
DEPARTMENT OF ELECTRONICS & TELECOMMUNICATION
ENGINEERING
PUNE INSTITUTE OF COMPUTER TECHNOLOGY
DHANKAWADI, PUNE- 411043
CERTIFICATE
This is to certify that the project report entitled
“ SMART GATE SUPERVISOR”
under the supervision of Dr. R. Sreemathy and it is approved for the partial fulfilment
of the requirement of Mini Project and Seminar subject of T.E. E&TC-2015 Course
of the Savitribai Phule Pune University, Pune.
Dr.R.Sreemathy Prof.Dr.Y.Ravinder
Project Guide HOD, E&TC Dept
Place: Pune
Date :
2
ACKNOWLEDGEMENT
The completion of this project could not have been possible without the
participation and assistance of so many people whose names may not all be
enumerated. Their contributions are sincerely appreciated and gratefully
acknowledged. However, the group would like to express their deep appreciation and
indebtedness particularly to the following:
Last but not the least, we would like to express our gratitude to our friends and
respondents for their support and willingness to be there with us, to share their
support, either morally, financially or physically.
3
INDEX
Abstract I
List of Figures II
List of Tables III
Chapter 1: Introduction
1.1 Background 9
1.2 Relevance. 9
1.3 Literature Survey 9
1.4 Motivation 10
1.5 Aim of the Project 10
1.6 Scope and Objectives 10
1.7 Technical Approach 11
4
Chapter 5: Results 23
References 25
Bill of Materials 26
5
ABSTRACT
The project is thought to be applicable for parking lots, automated toll booths
and law defaulters wherein the captured image of the vehicle can be processed to
obtain the license plate number and an uninterrupted surveillance can be installed.
The major processing step after obtaining a digital image in real time is
binarization of the image. The image will be then segmented and freed of noise, to get
the exact and proper vehicle number plate as the desired section of interest. Due to
binarization of the image, tariff vehicles having yellow license plates too can be
processed with equal efficiency.
The Extraction of characters will be done only of those vehicles which abide
by the standard font set by Regional Transport Office (RTO).
6
List of Figures:
7
List of Tables:
Sr.No Name
1 Comparison between Raspberry pi modules
2 Comparison between servo and stepper motor
3 Comparison between GSM SIM 800 and SIM 900
4 Bill of Materials
8
CHAPTER 1
INTRODUCTION
1.1 BACKGROUND:
1. Introduction/ Motivation :
Automatic recognition of car license plate number has become a
very important in our daily life because of the exponential increase of
cars and transportation systems which makes it impossible to be fully
managed and monitored by humans, examples are so many like traffic
monitoring, tracking stolen cars, managing parking toll, border and
customs checkpoints. Yet it’s a very challenging problem, due to the
diversity of plate formats, different scales, rotations and non-uniform
illumination conditions during image acquisition. Artificial neural is used
for character extraction and classification.
9
Security of people is the main concern whether it may be any
college, society or hospitals .As the population is increasing day by day,
there is a huge rush in the city and getting hold of checking each and
every vehicle while entering these premises is beyond the scope of 2-3
people around. Moreover a person can be bribed easily to create mishap
in the society .So we thought of making such automated systems
especially for the modern society and colleges so that these people are not
allowed to enter the premises.
One of the latest trends that will gain traction in the automatic number
plate recognition market in the coming years is the adoption of cloud-
based ANPR system. Cloud-based surveillance systems help end-users
benefit from numerous advantages such as pay-per-use, low initial
investment and maintenance, easy deployment, and instant access.
Additionally, the cloud-based model includes various services such as on-
and off-site storage, remote video monitoring, cloud-based video content
analytics, and real-time monitoring through the Internet.
10
An ANPR, or automated number plate recognition, system is an optical
character recognition system, which is implemented using cameras and
software to analyze the captured pictures. By following an extensive
research approach of qualitative and quantitative data collection methods
and by covering various factors including the growing demand for a
variety of products across geographical regions, emerging trends in
consumer behavior, prominent competitors, growth potential, and major
challenges; research analysts at the global market research
firm Technavio deliver a comprehensive collection of industry research
reports on the various aspects of the IT professional services, enterprise
applicationsand cloud computing sectors, which are an integral part of
the ICT sector industry.
11
3.2 Research Paper 2: An Efficient Method for Indian Vehicle License Plate
Extraction and Character Segmentation.
Summary:
This Research paper focuses on Character segmentation for license
plate recognition using RGB to Gray scale conversion methods. The author
has
testedthis approach in different illumination conditions and has got
satisfactory
results.
Objectives:
12
3. Study of Python Commands
13
CHAPTER 2
14
.
Explanation:
IR Sensor :Ir sensor is an electronic device that emits in order to sense some
aspects of the surroundings.In this project we are sensing the presence of
vehicle .As soon as vehicle arrives at the gate.Ir sensor detects its presence and
hence camera turns ON to click the image of car.
Web Camera: Any USB Digital Camera can be used. This is used to obtain a
digital image of the Vehicle.
15
2.3 ALGORITHM:
1. Start
2. Check whether a vehicle has arrived using IR sensor
3. Capture Image
4. Convert Image to Grayscale.
5. Convert Image into Binary (High Contrast Black and White).
6. Find all possible characters in the image
7. Apply Contour Detection.
8. Largest continuous Contour that has 4 Ends is the number plate. Select that
Contour.
9. Create Training Data for OCR(pre-requisite)
10. Apply OCR and extract digits and characters
11. Compare it with given dataset
12. Open gate if number matches
13. Go to step 1 and wait for next vehicle.
14. Fill the information on GUI.
15. Send the message containing information via GSM
16. Go to step 1 and wait for next vehicle
17. Display the status on TFT
16
FLOWCHART:
Start
Number plate
localization
Gray Scale
Binary
Extract Number
(OCR)
Match ?
END
Chapter 3
System Design
GSM :
Most of the people are familiarized with the dial up modem as it a common device for
land line telephone network used in homes and offices. It is a wired network which
18
has got its own limitations in the embedded area. GSM modem is introduced to rectify
the main limitation of the dial up modem based on its acceptance of a sim card. It is
almost equivalent to a mobile communication system as operates over a subscription
to a mobile operator. From the mobile operator perspective, a GSM modem looks just
like a mobile phone.Using the transmission and reception pins, a modem can receive
and send the messages and it could be interfaced with the PC or to a microcontroller.
This property makes the modem to exist in a relevant position on embedded
applications.
AT COMMANDS:
AT commands are instructions used to control a modem. AT is the abbreviation of
Attention. Every command line starts with “AT” or “at”. That’s why modem
commands are called AT commands.GSM modem support an AT command set that is
specific to the GSM technology, which include SMS-related commands like
AT+CMGS(send SMS message),AT+CMSS(send SMS message from
storage),ATE0(disable echo),AT+CMGF(select message format as text
mode),AT+CNMI(new SMS message indication)
.
Interfacing GSM with raspberry pi:
CONNECTION DIAGRAM:
19
Figure 2: Interfacing of GSM SIM 800 to Raspberry pi
Working of circuit:
For interfacing GSM to raspberry pi, following connections are made :
Rx of GSM to Tx of raspberry pi
Tx of GSM to rx of raspberry pi
GND to GND
As soon as AT command is being sent to the GSM, it sends an OK statement
as a response to the AT command being sent. As soon as OK statement is
being statement os being sent ,it ensures that the device is working properly.
And then various other commands like ATE0 is sent which disables the
ECHO. And as soon as AT+CMGS is being sent to GSM,it sends message to
the mobile no mentioned. The response of GSM modem to raspberry pi is
better than other processors/controllers.
IR SENSOR :
An infrared sensor is an electronic device, that emits in order to sense some aspects
of the surroundings. An IR sensor can measure the heat of an object as well as detects
the motion. These types of sensors measures only infrared radiation, rather than
emitting it that is called as a passive IR sensor. Usually in the infrared spectrum, all
the objects radiate some form of thermal radiations. These types of radiations are
invisible to our eyes, that can be detected by an infrared sensor. The emitter is simply
an IR LED (Light Emitting Diode) and the detector is simply an IR photodiode which
is sensitive to IR light of the same wavelength as that emitted by the IR LED. When
20
IR light falls on the photodiode, The resistances and these output voltages, change in
proportion to the magnitude of the IR light received.
IR SENSOR WORKING PRINCIPLE:
An infrared sensor is one of the basic and most popular circuit in the electronic era.IR
sensor is analogous to human’s visionary senses, which can be used to detect
obstacles and it is one of the common applications in real time.This circuit comprises
of the following components
PI CAM :
CONNECTION DIAGRAM :
22
Figure 5: Interfacing of Pi camera
DESCRIPTION :
This 8mp camera module is capable of 1080p video and still images that connect
directly to your Raspberry Pi. This is the plug-and-play-compatible latest version of
the Raspbian operating system, making it perfect for time-lapse photography,
recording video, motion detection and security applications. Connect the included
ribbon cable to the CSI (Camera Serial Interface) port on to Raspberry Pi.
The board itself is tiny, at around 25mm x 23mm x 9mm and weighing in at just over
3g, making it perfect for mobile or other applications where size and weight are
important. The sensor has a native resolution of 8 megapixel, and has a fixed focus
lens on board. In terms of still images, the camera is capable of 3280 x 2464 pixel
static images, and also supports 1080p30, 720p60 and 640x480p90 video.
Features :
SERVO MOTOR :
Servo implies an error sensing feedback control which is utilized to correct the
performance of a system. It also requires a generally sophisticated controller, often a
dedicated module designed particularly for use with servomotors. Servo motors are
23
DC motors that allows for precise control of angular position. They are actually DC
motors whose speed is slowly lowered by the gears. The servo motors usually have a
revolution cutoff from 90° to 180°.
The servo motor is actually an assembly of four things: a normal DC motor, a gear
reduction unit, a position-sensing device and a control circuit. DC motor is connected
with a gear mechanism which provides feedback to a position sensor which is mostly
a potentiometer. From the gear box, the output of the motor is delivered via servo
spline to the servo arm. For standard servo motors, the gear is normally made up of
plastic whereas for high power servos, the gear is made up of metal. The main
purpose of servo motor is to receive a control signal that represents a desired output
position of the servo shaft and apply power to its DC motor until its shaft turns to that
position.
Interface servo motor to raspberry pi
CONNECTION DIAGRAM
24
TFT :
TFT is short form for Thin Film Transistor .It is a type of LCD flat-panel display
screen in which each pixel is controlled by from one to four transistors. The TFT
technology provides the best resolution of all the flat panel techniques ,but it is also
the most expensive.TFT screens are sometimes called as active matrix LCDs.TFTs
are touch type LCDs.A display screen made with TFT (thin-film transistor)
technology is a liquid crystal display (LCD), common in notebook and laptop
computers, that has a transistor for each pixel (that is, for each of the tiny elements
that control the illumination of your display).Having a transistor at each pixel means
that the current that triggers pixel illumination can be smaller and therefore can be
switched on and off more quickly..
TFT is also known as active matrix display technology (and contrasts with "passive
matrix" which does not have a transistor at each pixel). A TFT or active matrix
display is more responsive to change. For example, when you move your mouse
across the screen, a TFT display is fast enough to reflect the movement of the mouse
cursor. (With a passive matrix display, the cursor temporarily disappears until the
display can "catch up.")
Features :
320×480 resolution
Resistive touch control
Supports any revision of Raspberry Pi (directly-pluggable)
Drivers provided (works with your own Raspbian/Ubuntu directly)
Size perfectly fits the Pi
High quality immersion gold surface plating
Supports Raspbian system, enables your system to:
25
Figure 7 :connection diagram of TFT
26
3.2SELECTION OF COMPONENTS:
USB Ports 2 4 4
27
Table 2: Comparison between SERVO and STEPPER Motors
Table 3: Comparison between GSM SIM 800 A and SIM 900 A Modules
28
CHAPTER 4
29
IMPLEMENTATION
Testing and Troubleshooting :
1. Installation of Raspberry pi
Problem :
Improper installation of raspberry pi due to corrupted image file.
Solution :
A new raspbean OS was installed with a proper disk image file.The
card got corrupted ,so replacing a new card solved the problem of
improper installation of raspbean OS.
2.OpenCV installation :
Problem :
The compilation of opencv used to error(race dependency condition
error ) out and the system hanged due to overload on system by the usage all 4
cores.
Solution :
Cleaning up of error installation and then using single core for
installation.
3. GSM :
Problem :
AT commands were not executed properly. GSM modem had to send an OK
statement after every AT commands but it could not do so.
Solution :
Serial port was not enabled .In raspberry pi 3,both bluetooth and UART are enabled
on the same port. There are basically two ways to communicate through raspberry pi
,ttyS0(internal UART enable),ttyAMA0(external device connection).So while
working on GSM module,one must disable ttyS0 and then connect external device to
it.
4. Pi Cam :
30
Problem : Error while capturing the image.
Solution : The camera need to be properly inserted in the slot and then camera must
be enabled. Once the camera is enabled ,raspberry pi booting must be done to ensure
no malfunctioning of pi cam.
4.2TESTING:
For Testing purposes and different angled photos of the car were taken so as to
observe whether the number plate was correctly located by the system, since light and
shadows play important part in profiling of image contours. The system shows quite a
good rate of fidelity since it gives correct output for 4 times out of 5 attempts.
CHAPTER 5
RESULT
31
Figure 8
32
Figure 9
Figure 10
33
Case 1:
Number plate is not present in the database
Figure 11
After entering all the data on GUI…a SMS will be sent to the concerned
authority through GSM modem interfaced with Raspberry pi
Screenshot of various messages sent are as shown below :
34
Figure 12
Case 2 :
Number plate is present in the database
If number plate is present in the database, a gate opening and closing mechanism is
inacted .
35
Figure 13
36
Figure 14
However the project lags a few points wherein the main shortcoming is the training of
OCR data. As for the training of data, KNN algorithm is being used , which is not as
efficient as Convolutional Neural Network .So the processing becomes sluggish
because of insufficient training of data.
Another major concern is the visual play of Lights, reflections and shadows under
different scenarios causes the location of number plate difficult , since the contour
formation is different under different light conditions.
The different processes shown above takes some time to execute as the data is being
trained everytime when the final code is being executed.
37
Control unit of Project:
38
Figure 15:
CHAPTER 6
39
FUTURE SCOPE AND CONCLUSION
FUTURE SCOPE:
CONCLUSION:
The Secure Access System can has widespread applications in real world and can help
personnel save valuable workforce by the use of this type of vehicular access security
systems.
The vast applications of Image Processing serves as an Electronic eye for man and
can robustly and continually provide surveillance as well as uncompromised and non
40
breachable security. Tools like Open Source Computer Vision Library Provide a good
platform for implementing Image Processing Activity.
REFERENCES
[2] Kumar Parsuraman "An Efficient Method for Indian Vehicle License Plate
Extraction and Character Segmentation" 2010 IEEE International Conference on
computational intelligence and computing research
41
Table no 3:List of Expenses
TOTAL 6195.00
42