SlideShare a Scribd company logo
Page | 1
Project Report
ADAPTIVE AND AUTONOMOUS TRAFFIC MANAGEMENT SYSTEM
(AATMS)
Team ID: NIY-63
Team Mentor: Amit Yadav
Institution: Delhi Technological University (DTU)
Team Members: Nitish Goel, Praneet Soni, Parth Sood, Mukul Aggarwal, Prakhar Agarwal
INTRODUCTION:
The Adaptive and Autonomous Traffic Management System (AATMS) is an Intelligent
Transport System(ITS) and is of considerable interest because of its potential in managing real-time
multi- lane traffic using a camera interfaced with a microcontroller. The project introduces novel
framework for density/number of cars calculation, license plate detection and Ambulance detection.
APPLICATION:
The purpose of this paper is to develop a real-time dynamic traffic management system that 1)
maintains the indigenous and optimized flow of traffic by calculating the density/number of vehicles
in several lanes using standard image processing techniques, 2) recognizes the license plates of the
vehicles using OCR Algorithm and automatically generates their database so that defaulters can be
identified and 3) creates a delay-free corridor for Ambulance(Emergency Vehicle) by identifying it
in a particular lane using Sound Processing techniques and giving green signal all the way through.
The project aims at improving efficiency, safety and minimization of human intervention.
SOFTWARE USED: NI LabView 2013, NI Vision Development Module 2013,
NI Vision Acquisition Express 2013
HARDWARE USED: Led's(Red and Green), NI myRIO, Webcams, Powered USB Hub,
RF Module(434MHz), Switches
DETAILS OF THE PROJECT:
The tasks implemented in the project are:-
 Density calculation/number of vehicles detection
 License plate detection and identification
 Ambulance detection
1) Density/Number of vehicles detection
We have used a model of a 4-way intersection and dummy cars for working illustration. The input is
taken from webcams and the output is presented through led's (1 red led, 1 green led per traffic light)
connected to digital I/O pins of myRIO.
We have also successfully implemented the code on recorded real-time traffic feed from 4 different
intersections which shows the robustness and real time implementation of our project.
The above task is divided into two different subtasks
i. Detection of vehicles in each lane
ii. Traffic lights control algorithm
Page | 2
i) Detection of vehicles in each lane: Vehicle detection in performed using Background Subtraction
followed by image processing techniques such as thresholding, morphology and particle analysis.
The following code is implemented for each one of the four lanes.
 The background image(reference frame) and the current frame are acquired through Vision
Acquisition Express VI and are converted to grayscale using Color Plane Extraction in
Vision Assistant Express VI.
 The reference frame is subtracted from the current frame using IMAQ Absolute Difference
VI.
 The number of cars are detected after series of image processing techniques in Vision Assistant
Express VI VDM Final3.
Fig: Processing Functions of VDM Final3
VDM Fnal3 VI
1) Threshold1 is used to filter out the pixels of the foreground objects on the basis of the threshold
value.
2) Series of morphological operations are performed to remove the unwanted particles.
3) Particle Analysis calculates the number and the area of the detected objects (in pixels).
(a) (b)
Page | 3
2) Left turn is always free
3) The duration of forward green signal and right
turn green signal is equal.
In the given figure, during one iteration, Lane 4 has the
active green signal while other three lanes have active
red signal.(T.L.=Traffic Light)
The code for the algorithm is given below:-
(c) (d)
Fig: (a) Background Model; (b) Current Frame
(c) Subtracted Image;
(d) After series of morphology operations
(e) Particle Analysis (e)
ii) Traffic lights control algorithm: The code iterates
cyclically through Lane 1 to Lane 4 and controls the
duration of green signal depending upon the number of
vehicles in that lane. We have taken the assumption 1) that
at a time, only one lane has the active green signal
1
2 3
4
T.L
.
Page | 4
Every iteration of the while loop involves two sequential steps:
Step 1:
Each of the camera returns the number of cars in the respective lane. All these values are
combined into a bundle which is then converted into a 1D array. Each of the these values is
accessed sequentially starting from index 0 to 3.
The value of index to be accessed is decided by the remainder obtained on division of i (iteration
terminal) by 4. In this way, the value iterates from 0 to 3 and then back to 3.
The element accessed is now subjected to 3 range tests, which help in classifying the
corresponding lane into one of the three categories defined, on the basis of number of cars
detected. After the classification, the corresponding wait time is assigned to the green signal of the
that lane.
Step 2:
The green signal of the selected lane is activated and the program waits for the specified time.
After the wait is complete, the while loop iterates again but the lane selected now is the one
in sequence with previous lane.
Also we have introduced and Override Switch. When the switch is active, each lane is assigned an
equal timing for active green signal. This idea can be used when the cameras have to be uninstalled
for maintenance or any other external factors.
RESULTS: (For the 30th Frame)
Page | 5
Page | 6
2) License plate detection and recognition: The license plate recognition (LPR) system proposed is
based on two main phases: License plate localization, and Segmentation with OCR.
As an initial step, the license plate must be located in order to read the characters. After locating
the plate, masking is performed to extract only the license plate region out of the whole image.
The extracted portion, which is the license plate, is then forwarded to the segmentation and OCR
phase.
The image below gives the LABVIEW Block Diagram for the proposed system.
Fig.1: LABVIEW Block Diagram of the proposed system.
 Both the Vision Acquisition Express VIs are used to acquire the image of the vehicle.
 The first Vision Assistant Express VI extracts the ROI consisting of only the license plate
characters. It returns a mask for this purpose.
 The mask from the above Vision Assistant is converted to ROI with the use of “IMAQ
MaskToROI”.
 The ROI obtained is used to extract the characters of the license plate in the second Vision
Assistant. OCR is then applied on the segmented characters.
License Plate Localization
The license plate localization phase is a very crucial stage. Failing to achieve this task will not
allow the reading of the plate to proceed. The following subsections explain in detail the two main
phases of the system.
Fig.2: Processing Functions used in the ROI Extractor Vision Assistant Express VI.
Page | 7
ROI EXTRACTOR EXPRESS VI
 Color plane extraction is used to extract the intensity plane.
 Auto Threshold: Clustering is used to look for bright objects.
 Binary Image Inversion is performed.
 Series of Morphological operations are performed to remove the unwanted particles from
the image.
 Finally, Image mask is used to extract the license plate region based on the approx. location
of the number plate in the image.
(a) (b)
(c) (d)
Fig.3: (a) Original Image; (b) After binary image inversion; (c) After morphological operations;
(d) After image mask.
Segmentation with OCR
OCR is the process by which the system reads characters in an image after separating them in
blocks. The separation process is called Segmentation.
Fig.4: Processing Functions used in the OCR Vision Assistant Express VI.
OCR EXPRESS VI
 Image mask from the previous VI is converted into ROI and applied on the original image
of the vehicle.
 Color plane extraction and threshold is applied on the masked ROI. The threshold method
used is Inter-Variance.
 Binary image inversion is performed and border objects are removed.
 Particle filter is used to remove the particles with smaller area as compared to the
characters.
 OCR can now be applied as only the characters are present in the final image.
Page | 8
(b)
(c)
(d)
(a)
(e)
Fig.5: (a) Original Image; (b) Masked Image; (c) Threshold Image; (d)After Particle filter;
(e) Final OCR.
RESULTS:
Page | 9
The above code is implemented on still images for working illustration. We have implemented the
algorithm on real time videos which shows the practical implementation of our method.
Page | 10
3) Ambulance detection: The Ambulance detection is performed in two steps:
 Detection of presence of Ambulance in one of the four lanes
 Identification of that lane
i) Detection of presence of Ambulance in one of the four lanes:
The presence of Ambulance is detected using Sound Processing techniques. It involves the analysis
of the spectrum of the Fast Fourier Transform(fft) of the sound of the siren of the Ambulance.
The following code detected the sound of the siren of the Ambulance:
Page | 11
 The ambulance siren oscillates between two dominant frequencies, the code exploits this
acoustic characteristic of the siren in order to identify it.
 First of all the input signal is passed through a band pass filter and signal between 800khz and
1500khz is extracted, because dominant frequencies of siren usually lie in this range. We are
not interested in other frequency components.
 Then we find 5 nodes from this signal and calculate the mean to find approximately the
frequency about which the spectrum is spread.
 If mean is <900khz, the given signal is not the siren and no further processing is done.
 If mean is >900khz , then the signal is further processed.
 The fft of the original signal is obtained and the spectrum is divided into 2 parts: 800khz to
1000khz and 1000khz to 1500khz. The peak is found in each of the two segments.
 The frequency at which the peak occurs and the amplitude at that frequency is checked. If they
are in the prescribed range in both the segments ,then the signal is the siren and the output led
is turned on.
RESULTS:
i) Identification of that lane:
We are using RF modules, working at 434Mhz frequency for the detection of the Ambulance
Lane. The driver in the lane will transmit the lane number to the RF receiver, which is connected
to myRIO . We have used 8-bit address lines as the encryption mechanism so that only the
authorized vehicles can transmit the valid signal. When the driver selects a particular lane the
address lines at the transmitter acquires a unique 8-bit value. Data at the receiver is valid , only
when address pins at receiver and transmitter are same. So there are 4 different 8-bit values of the
address line , each corresponding to one lane. The address lines at the transmitter are achieved
through hardware connections.
Page | 12
At the receiver, myRIO sequentially assigns the 4 8-bit addresses to the receiver address lines. So
the receiver is identified only when the received address lines matches any one of the four
assigned address lines.
8 Bit
9V Battery
(RF Communication)
D0 D1 D2
D0
D1
CONTROL BOARD Data at receiver is
valid if address pins of
transmitter-receiver match
D1
D0
D0
D1
D1
D0
1
D0
A
D
D
R
E
S
S
L
I
N
E
S
Vcc
RF
TRANSMITTER
RF
RECIEVER
Vcc
LANE CONTROL
SWITCH
ENABLE
D0 D1 D2
myRIO
O/p
8 pins to led's of 4 lanes
8-Bit
Address
Lines
Page | 13
RESULTS:-

More Related Content

What's hot

Vehicle detection through image processing
Vehicle detection through image processingVehicle detection through image processing
Vehicle detection through image processingGhazalpreet Kaur
 
Identification and classification of moving vehicles on road
Identification and classification of moving vehicles on roadIdentification and classification of moving vehicles on road
Identification and classification of moving vehicles on road
Alexander Decker
 
automatic number plate recognition
automatic number plate recognitionautomatic number plate recognition
automatic number plate recognition
Sairam Taduvai
 
Auto Traffic Management System
Auto Traffic Management SystemAuto Traffic Management System
Auto Traffic Management System
vivatechijri
 
Automatic vehicle license plate detection using VEDA
Automatic vehicle license plate detection using VEDAAutomatic vehicle license plate detection using VEDA
Automatic vehicle license plate detection using VEDA
Rojith Thomas
 
A0140109
A0140109A0140109
A0140109
IOSR Journals
 
Computer Vision for Traffic Sign Recognition
Computer Vision for Traffic Sign RecognitionComputer Vision for Traffic Sign Recognition
Computer Vision for Traffic Sign Recognitionthevijayps
 
Automatic detection and recognition of road sign for driver assistance system
Automatic detection and recognition of road sign for driver assistance systemAutomatic detection and recognition of road sign for driver assistance system
Automatic detection and recognition of road sign for driver assistance system
sudhakar5472
 
Automatic Road Sign Recognition From Video
Automatic Road Sign Recognition From VideoAutomatic Road Sign Recognition From Video
Automatic Road Sign Recognition From Video
Dr Wei Liu
 
IRJET - Autonomous Eviscerating BOT using ANT Colony Optimization
IRJET -  	  Autonomous Eviscerating BOT using ANT Colony OptimizationIRJET -  	  Autonomous Eviscerating BOT using ANT Colony Optimization
IRJET - Autonomous Eviscerating BOT using ANT Colony Optimization
IRJET Journal
 
LICENSE NUMBER PLATE RECOGNITION SYSTEM USING ANDROID APP
LICENSE NUMBER PLATE RECOGNITION SYSTEM USING ANDROID APPLICENSE NUMBER PLATE RECOGNITION SYSTEM USING ANDROID APP
LICENSE NUMBER PLATE RECOGNITION SYSTEM USING ANDROID APP
Aditya Mishra
 
Image Processing Applied To Traffic Queue Detection Algorithm
Image Processing Applied To Traffic Queue Detection AlgorithmImage Processing Applied To Traffic Queue Detection Algorithm
Image Processing Applied To Traffic Queue Detection Algorithmguest673189
 
Autonomous Traffic Signal Control using Decision Tree
Autonomous Traffic Signal Control using Decision Tree Autonomous Traffic Signal Control using Decision Tree
Autonomous Traffic Signal Control using Decision Tree
IJECEIAES
 
IRJET- Intelligent Queue Management System at Airports using Image Processing...
IRJET- Intelligent Queue Management System at Airports using Image Processing...IRJET- Intelligent Queue Management System at Airports using Image Processing...
IRJET- Intelligent Queue Management System at Airports using Image Processing...
IRJET Journal
 
IRJET- Video Based Traffic Sign Detection by Scale Based Frame Fusion Technique
IRJET- Video Based Traffic Sign Detection by Scale Based Frame Fusion TechniqueIRJET- Video Based Traffic Sign Detection by Scale Based Frame Fusion Technique
IRJET- Video Based Traffic Sign Detection by Scale Based Frame Fusion Technique
IRJET Journal
 
TRAFFIC-SIGN RECOGNITION FOR AN INTELLIGENT VEHICLE/DRIVER ASSISTANT SYSTEM U...
TRAFFIC-SIGN RECOGNITION FOR AN INTELLIGENT VEHICLE/DRIVER ASSISTANT SYSTEM U...TRAFFIC-SIGN RECOGNITION FOR AN INTELLIGENT VEHICLE/DRIVER ASSISTANT SYSTEM U...
TRAFFIC-SIGN RECOGNITION FOR AN INTELLIGENT VEHICLE/DRIVER ASSISTANT SYSTEM U...
cseij
 
Automatic License Plate Recognition using OpenCV
Automatic License Plate Recognition using OpenCVAutomatic License Plate Recognition using OpenCV
Automatic License Plate Recognition using OpenCV
Editor IJCATR
 
AN EFFICIENT SYSTEM FOR FORWARD COLLISION AVOIDANCE USING LOW COST CAMERA & E...
AN EFFICIENT SYSTEM FOR FORWARD COLLISION AVOIDANCE USING LOW COST CAMERA & E...AN EFFICIENT SYSTEM FOR FORWARD COLLISION AVOIDANCE USING LOW COST CAMERA & E...
AN EFFICIENT SYSTEM FOR FORWARD COLLISION AVOIDANCE USING LOW COST CAMERA & E...
aciijournal
 
A Vision based Driver Support System for Road Sign Detection
A Vision based Driver Support System for Road Sign DetectionA Vision based Driver Support System for Road Sign Detection
A Vision based Driver Support System for Road Sign Detection
idescitation
 

What's hot (20)

Vehicle detection through image processing
Vehicle detection through image processingVehicle detection through image processing
Vehicle detection through image processing
 
Identification and classification of moving vehicles on road
Identification and classification of moving vehicles on roadIdentification and classification of moving vehicles on road
Identification and classification of moving vehicles on road
 
automatic number plate recognition
automatic number plate recognitionautomatic number plate recognition
automatic number plate recognition
 
Auto Traffic Management System
Auto Traffic Management SystemAuto Traffic Management System
Auto Traffic Management System
 
Automatic vehicle license plate detection using VEDA
Automatic vehicle license plate detection using VEDAAutomatic vehicle license plate detection using VEDA
Automatic vehicle license plate detection using VEDA
 
A0140109
A0140109A0140109
A0140109
 
Computer Vision for Traffic Sign Recognition
Computer Vision for Traffic Sign RecognitionComputer Vision for Traffic Sign Recognition
Computer Vision for Traffic Sign Recognition
 
Automatic detection and recognition of road sign for driver assistance system
Automatic detection and recognition of road sign for driver assistance systemAutomatic detection and recognition of road sign for driver assistance system
Automatic detection and recognition of road sign for driver assistance system
 
Automatic Road Sign Recognition From Video
Automatic Road Sign Recognition From VideoAutomatic Road Sign Recognition From Video
Automatic Road Sign Recognition From Video
 
JMS_Luo
JMS_LuoJMS_Luo
JMS_Luo
 
IRJET - Autonomous Eviscerating BOT using ANT Colony Optimization
IRJET -  	  Autonomous Eviscerating BOT using ANT Colony OptimizationIRJET -  	  Autonomous Eviscerating BOT using ANT Colony Optimization
IRJET - Autonomous Eviscerating BOT using ANT Colony Optimization
 
LICENSE NUMBER PLATE RECOGNITION SYSTEM USING ANDROID APP
LICENSE NUMBER PLATE RECOGNITION SYSTEM USING ANDROID APPLICENSE NUMBER PLATE RECOGNITION SYSTEM USING ANDROID APP
LICENSE NUMBER PLATE RECOGNITION SYSTEM USING ANDROID APP
 
Image Processing Applied To Traffic Queue Detection Algorithm
Image Processing Applied To Traffic Queue Detection AlgorithmImage Processing Applied To Traffic Queue Detection Algorithm
Image Processing Applied To Traffic Queue Detection Algorithm
 
Autonomous Traffic Signal Control using Decision Tree
Autonomous Traffic Signal Control using Decision Tree Autonomous Traffic Signal Control using Decision Tree
Autonomous Traffic Signal Control using Decision Tree
 
IRJET- Intelligent Queue Management System at Airports using Image Processing...
IRJET- Intelligent Queue Management System at Airports using Image Processing...IRJET- Intelligent Queue Management System at Airports using Image Processing...
IRJET- Intelligent Queue Management System at Airports using Image Processing...
 
IRJET- Video Based Traffic Sign Detection by Scale Based Frame Fusion Technique
IRJET- Video Based Traffic Sign Detection by Scale Based Frame Fusion TechniqueIRJET- Video Based Traffic Sign Detection by Scale Based Frame Fusion Technique
IRJET- Video Based Traffic Sign Detection by Scale Based Frame Fusion Technique
 
TRAFFIC-SIGN RECOGNITION FOR AN INTELLIGENT VEHICLE/DRIVER ASSISTANT SYSTEM U...
TRAFFIC-SIGN RECOGNITION FOR AN INTELLIGENT VEHICLE/DRIVER ASSISTANT SYSTEM U...TRAFFIC-SIGN RECOGNITION FOR AN INTELLIGENT VEHICLE/DRIVER ASSISTANT SYSTEM U...
TRAFFIC-SIGN RECOGNITION FOR AN INTELLIGENT VEHICLE/DRIVER ASSISTANT SYSTEM U...
 
Automatic License Plate Recognition using OpenCV
Automatic License Plate Recognition using OpenCVAutomatic License Plate Recognition using OpenCV
Automatic License Plate Recognition using OpenCV
 
AN EFFICIENT SYSTEM FOR FORWARD COLLISION AVOIDANCE USING LOW COST CAMERA & E...
AN EFFICIENT SYSTEM FOR FORWARD COLLISION AVOIDANCE USING LOW COST CAMERA & E...AN EFFICIENT SYSTEM FOR FORWARD COLLISION AVOIDANCE USING LOW COST CAMERA & E...
AN EFFICIENT SYSTEM FOR FORWARD COLLISION AVOIDANCE USING LOW COST CAMERA & E...
 
A Vision based Driver Support System for Road Sign Detection
A Vision based Driver Support System for Road Sign DetectionA Vision based Driver Support System for Road Sign Detection
A Vision based Driver Support System for Road Sign Detection
 

Similar to Report NIYANTRA

Automatic Vehicle Detection Using Pixelwise Classification Approach
Automatic Vehicle Detection Using Pixelwise Classification ApproachAutomatic Vehicle Detection Using Pixelwise Classification Approach
Automatic Vehicle Detection Using Pixelwise Classification Approach
IOSR Journals
 
Real time-image-processing-applied-to-traffic-queue-detection-algorithm
Real time-image-processing-applied-to-traffic-queue-detection-algorithmReal time-image-processing-applied-to-traffic-queue-detection-algorithm
Real time-image-processing-applied-to-traffic-queue-detection-algorithmajayrampelli
 
traffic jam detection using image processing
traffic jam detection using image processingtraffic jam detection using image processing
traffic jam detection using image processingMalika Alix
 
Traffic Light Detection and Recognition for Self Driving Cars using Deep Lear...
Traffic Light Detection and Recognition for Self Driving Cars using Deep Lear...Traffic Light Detection and Recognition for Self Driving Cars using Deep Lear...
Traffic Light Detection and Recognition for Self Driving Cars using Deep Lear...
ijtsrd
 
License Plate Recognition
License Plate RecognitionLicense Plate Recognition
License Plate Recognition
Amr Rashed
 
Smart Control of Traffic Signal System using Image Processing
Smart Control of Traffic Signal System using Image Processing Smart Control of Traffic Signal System using Image Processing
Smart Control of Traffic Signal System using Image Processing
Raihan Bin-Mofidul
 
Vehicle counting for traffic management
Vehicle counting for traffic management Vehicle counting for traffic management
Vehicle counting for traffic management
ADEEBANADEEM
 
A Study on Single Camera Based ANPR System for Improvement of Vehicle Number ...
A Study on Single Camera Based ANPR System for Improvement of Vehicle Number ...A Study on Single Camera Based ANPR System for Improvement of Vehicle Number ...
A Study on Single Camera Based ANPR System for Improvement of Vehicle Number ...
IJRTEMJOURNAL
 
QDA_RTP_Traffic_ppt_final.ppt
QDA_RTP_Traffic_ppt_final.pptQDA_RTP_Traffic_ppt_final.ppt
QDA_RTP_Traffic_ppt_final.ppt
PRATAP'S MOBILE ARENA
 
Automatic number plate recognition (anpr)
Automatic number plate recognition (anpr)Automatic number plate recognition (anpr)
Automatic number plate recognition (anpr)
AbhishekChoudhary464889
 
Neural Network based Vehicle Classification for Intelligent Traffic Control
Neural Network based Vehicle Classification for Intelligent Traffic ControlNeural Network based Vehicle Classification for Intelligent Traffic Control
Neural Network based Vehicle Classification for Intelligent Traffic Control
ijseajournal
 
An Efficient Model to Identify A Vehicle by Recognizing the Alphanumeric Char...
An Efficient Model to Identify A Vehicle by Recognizing the Alphanumeric Char...An Efficient Model to Identify A Vehicle by Recognizing the Alphanumeric Char...
An Efficient Model to Identify A Vehicle by Recognizing the Alphanumeric Char...
IJMTST Journal
 
Smart Algorithm for Traffic Congestion and Control
Smart  Algorithm for Traffic Congestion and ControlSmart  Algorithm for Traffic Congestion and Control
Smart Algorithm for Traffic Congestion and Control
IRJET Journal
 
A real-time system for vehicle detection with shadow removal and vehicle clas...
A real-time system for vehicle detection with shadow removal and vehicle clas...A real-time system for vehicle detection with shadow removal and vehicle clas...
A real-time system for vehicle detection with shadow removal and vehicle clas...
International Journal of Power Electronics and Drive Systems
 
Automatic Fetching of Vehicle details using ANPR Camera
Automatic Fetching of Vehicle details using ANPR CameraAutomatic Fetching of Vehicle details using ANPR Camera
Automatic Fetching of Vehicle details using ANPR Camera
IRJET Journal
 
Color Tracking Robot
Color Tracking RobotColor Tracking Robot
Color Tracking Robot
paperpublications3
 
Mobile Based Application to Scan the Number Plate and To Verify the Owner Det...
Mobile Based Application to Scan the Number Plate and To Verify the Owner Det...Mobile Based Application to Scan the Number Plate and To Verify the Owner Det...
Mobile Based Application to Scan the Number Plate and To Verify the Owner Det...
inventionjournals
 
IRJET- Dynamic Traffic Management System
IRJET- Dynamic Traffic Management SystemIRJET- Dynamic Traffic Management System
IRJET- Dynamic Traffic Management System
IRJET Journal
 
IRJET- Time To Cross – Traffic Light Control System using Image Processing
IRJET-  	  Time To Cross – Traffic Light Control System using Image ProcessingIRJET-  	  Time To Cross – Traffic Light Control System using Image Processing
IRJET- Time To Cross – Traffic Light Control System using Image Processing
IRJET Journal
 
License plate extraction of overspeeding vehicles
License plate extraction of overspeeding vehiclesLicense plate extraction of overspeeding vehicles
License plate extraction of overspeeding vehicles
lambanaveen
 

Similar to Report NIYANTRA (20)

Automatic Vehicle Detection Using Pixelwise Classification Approach
Automatic Vehicle Detection Using Pixelwise Classification ApproachAutomatic Vehicle Detection Using Pixelwise Classification Approach
Automatic Vehicle Detection Using Pixelwise Classification Approach
 
Real time-image-processing-applied-to-traffic-queue-detection-algorithm
Real time-image-processing-applied-to-traffic-queue-detection-algorithmReal time-image-processing-applied-to-traffic-queue-detection-algorithm
Real time-image-processing-applied-to-traffic-queue-detection-algorithm
 
traffic jam detection using image processing
traffic jam detection using image processingtraffic jam detection using image processing
traffic jam detection using image processing
 
Traffic Light Detection and Recognition for Self Driving Cars using Deep Lear...
Traffic Light Detection and Recognition for Self Driving Cars using Deep Lear...Traffic Light Detection and Recognition for Self Driving Cars using Deep Lear...
Traffic Light Detection and Recognition for Self Driving Cars using Deep Lear...
 
License Plate Recognition
License Plate RecognitionLicense Plate Recognition
License Plate Recognition
 
Smart Control of Traffic Signal System using Image Processing
Smart Control of Traffic Signal System using Image Processing Smart Control of Traffic Signal System using Image Processing
Smart Control of Traffic Signal System using Image Processing
 
Vehicle counting for traffic management
Vehicle counting for traffic management Vehicle counting for traffic management
Vehicle counting for traffic management
 
A Study on Single Camera Based ANPR System for Improvement of Vehicle Number ...
A Study on Single Camera Based ANPR System for Improvement of Vehicle Number ...A Study on Single Camera Based ANPR System for Improvement of Vehicle Number ...
A Study on Single Camera Based ANPR System for Improvement of Vehicle Number ...
 
QDA_RTP_Traffic_ppt_final.ppt
QDA_RTP_Traffic_ppt_final.pptQDA_RTP_Traffic_ppt_final.ppt
QDA_RTP_Traffic_ppt_final.ppt
 
Automatic number plate recognition (anpr)
Automatic number plate recognition (anpr)Automatic number plate recognition (anpr)
Automatic number plate recognition (anpr)
 
Neural Network based Vehicle Classification for Intelligent Traffic Control
Neural Network based Vehicle Classification for Intelligent Traffic ControlNeural Network based Vehicle Classification for Intelligent Traffic Control
Neural Network based Vehicle Classification for Intelligent Traffic Control
 
An Efficient Model to Identify A Vehicle by Recognizing the Alphanumeric Char...
An Efficient Model to Identify A Vehicle by Recognizing the Alphanumeric Char...An Efficient Model to Identify A Vehicle by Recognizing the Alphanumeric Char...
An Efficient Model to Identify A Vehicle by Recognizing the Alphanumeric Char...
 
Smart Algorithm for Traffic Congestion and Control
Smart  Algorithm for Traffic Congestion and ControlSmart  Algorithm for Traffic Congestion and Control
Smart Algorithm for Traffic Congestion and Control
 
A real-time system for vehicle detection with shadow removal and vehicle clas...
A real-time system for vehicle detection with shadow removal and vehicle clas...A real-time system for vehicle detection with shadow removal and vehicle clas...
A real-time system for vehicle detection with shadow removal and vehicle clas...
 
Automatic Fetching of Vehicle details using ANPR Camera
Automatic Fetching of Vehicle details using ANPR CameraAutomatic Fetching of Vehicle details using ANPR Camera
Automatic Fetching of Vehicle details using ANPR Camera
 
Color Tracking Robot
Color Tracking RobotColor Tracking Robot
Color Tracking Robot
 
Mobile Based Application to Scan the Number Plate and To Verify the Owner Det...
Mobile Based Application to Scan the Number Plate and To Verify the Owner Det...Mobile Based Application to Scan the Number Plate and To Verify the Owner Det...
Mobile Based Application to Scan the Number Plate and To Verify the Owner Det...
 
IRJET- Dynamic Traffic Management System
IRJET- Dynamic Traffic Management SystemIRJET- Dynamic Traffic Management System
IRJET- Dynamic Traffic Management System
 
IRJET- Time To Cross – Traffic Light Control System using Image Processing
IRJET-  	  Time To Cross – Traffic Light Control System using Image ProcessingIRJET-  	  Time To Cross – Traffic Light Control System using Image Processing
IRJET- Time To Cross – Traffic Light Control System using Image Processing
 
License plate extraction of overspeeding vehicles
License plate extraction of overspeeding vehiclesLicense plate extraction of overspeeding vehicles
License plate extraction of overspeeding vehicles
 

Report NIYANTRA

  • 1. Page | 1 Project Report ADAPTIVE AND AUTONOMOUS TRAFFIC MANAGEMENT SYSTEM (AATMS) Team ID: NIY-63 Team Mentor: Amit Yadav Institution: Delhi Technological University (DTU) Team Members: Nitish Goel, Praneet Soni, Parth Sood, Mukul Aggarwal, Prakhar Agarwal INTRODUCTION: The Adaptive and Autonomous Traffic Management System (AATMS) is an Intelligent Transport System(ITS) and is of considerable interest because of its potential in managing real-time multi- lane traffic using a camera interfaced with a microcontroller. The project introduces novel framework for density/number of cars calculation, license plate detection and Ambulance detection. APPLICATION: The purpose of this paper is to develop a real-time dynamic traffic management system that 1) maintains the indigenous and optimized flow of traffic by calculating the density/number of vehicles in several lanes using standard image processing techniques, 2) recognizes the license plates of the vehicles using OCR Algorithm and automatically generates their database so that defaulters can be identified and 3) creates a delay-free corridor for Ambulance(Emergency Vehicle) by identifying it in a particular lane using Sound Processing techniques and giving green signal all the way through. The project aims at improving efficiency, safety and minimization of human intervention. SOFTWARE USED: NI LabView 2013, NI Vision Development Module 2013, NI Vision Acquisition Express 2013 HARDWARE USED: Led's(Red and Green), NI myRIO, Webcams, Powered USB Hub, RF Module(434MHz), Switches DETAILS OF THE PROJECT: The tasks implemented in the project are:-  Density calculation/number of vehicles detection  License plate detection and identification  Ambulance detection 1) Density/Number of vehicles detection We have used a model of a 4-way intersection and dummy cars for working illustration. The input is taken from webcams and the output is presented through led's (1 red led, 1 green led per traffic light) connected to digital I/O pins of myRIO. We have also successfully implemented the code on recorded real-time traffic feed from 4 different intersections which shows the robustness and real time implementation of our project. The above task is divided into two different subtasks i. Detection of vehicles in each lane ii. Traffic lights control algorithm
  • 2. Page | 2 i) Detection of vehicles in each lane: Vehicle detection in performed using Background Subtraction followed by image processing techniques such as thresholding, morphology and particle analysis. The following code is implemented for each one of the four lanes.  The background image(reference frame) and the current frame are acquired through Vision Acquisition Express VI and are converted to grayscale using Color Plane Extraction in Vision Assistant Express VI.  The reference frame is subtracted from the current frame using IMAQ Absolute Difference VI.  The number of cars are detected after series of image processing techniques in Vision Assistant Express VI VDM Final3. Fig: Processing Functions of VDM Final3 VDM Fnal3 VI 1) Threshold1 is used to filter out the pixels of the foreground objects on the basis of the threshold value. 2) Series of morphological operations are performed to remove the unwanted particles. 3) Particle Analysis calculates the number and the area of the detected objects (in pixels). (a) (b)
  • 3. Page | 3 2) Left turn is always free 3) The duration of forward green signal and right turn green signal is equal. In the given figure, during one iteration, Lane 4 has the active green signal while other three lanes have active red signal.(T.L.=Traffic Light) The code for the algorithm is given below:- (c) (d) Fig: (a) Background Model; (b) Current Frame (c) Subtracted Image; (d) After series of morphology operations (e) Particle Analysis (e) ii) Traffic lights control algorithm: The code iterates cyclically through Lane 1 to Lane 4 and controls the duration of green signal depending upon the number of vehicles in that lane. We have taken the assumption 1) that at a time, only one lane has the active green signal 1 2 3 4 T.L .
  • 4. Page | 4 Every iteration of the while loop involves two sequential steps: Step 1: Each of the camera returns the number of cars in the respective lane. All these values are combined into a bundle which is then converted into a 1D array. Each of the these values is accessed sequentially starting from index 0 to 3. The value of index to be accessed is decided by the remainder obtained on division of i (iteration terminal) by 4. In this way, the value iterates from 0 to 3 and then back to 3. The element accessed is now subjected to 3 range tests, which help in classifying the corresponding lane into one of the three categories defined, on the basis of number of cars detected. After the classification, the corresponding wait time is assigned to the green signal of the that lane. Step 2: The green signal of the selected lane is activated and the program waits for the specified time. After the wait is complete, the while loop iterates again but the lane selected now is the one in sequence with previous lane. Also we have introduced and Override Switch. When the switch is active, each lane is assigned an equal timing for active green signal. This idea can be used when the cameras have to be uninstalled for maintenance or any other external factors. RESULTS: (For the 30th Frame)
  • 6. Page | 6 2) License plate detection and recognition: The license plate recognition (LPR) system proposed is based on two main phases: License plate localization, and Segmentation with OCR. As an initial step, the license plate must be located in order to read the characters. After locating the plate, masking is performed to extract only the license plate region out of the whole image. The extracted portion, which is the license plate, is then forwarded to the segmentation and OCR phase. The image below gives the LABVIEW Block Diagram for the proposed system. Fig.1: LABVIEW Block Diagram of the proposed system.  Both the Vision Acquisition Express VIs are used to acquire the image of the vehicle.  The first Vision Assistant Express VI extracts the ROI consisting of only the license plate characters. It returns a mask for this purpose.  The mask from the above Vision Assistant is converted to ROI with the use of “IMAQ MaskToROI”.  The ROI obtained is used to extract the characters of the license plate in the second Vision Assistant. OCR is then applied on the segmented characters. License Plate Localization The license plate localization phase is a very crucial stage. Failing to achieve this task will not allow the reading of the plate to proceed. The following subsections explain in detail the two main phases of the system. Fig.2: Processing Functions used in the ROI Extractor Vision Assistant Express VI.
  • 7. Page | 7 ROI EXTRACTOR EXPRESS VI  Color plane extraction is used to extract the intensity plane.  Auto Threshold: Clustering is used to look for bright objects.  Binary Image Inversion is performed.  Series of Morphological operations are performed to remove the unwanted particles from the image.  Finally, Image mask is used to extract the license plate region based on the approx. location of the number plate in the image. (a) (b) (c) (d) Fig.3: (a) Original Image; (b) After binary image inversion; (c) After morphological operations; (d) After image mask. Segmentation with OCR OCR is the process by which the system reads characters in an image after separating them in blocks. The separation process is called Segmentation. Fig.4: Processing Functions used in the OCR Vision Assistant Express VI. OCR EXPRESS VI  Image mask from the previous VI is converted into ROI and applied on the original image of the vehicle.  Color plane extraction and threshold is applied on the masked ROI. The threshold method used is Inter-Variance.  Binary image inversion is performed and border objects are removed.  Particle filter is used to remove the particles with smaller area as compared to the characters.  OCR can now be applied as only the characters are present in the final image.
  • 8. Page | 8 (b) (c) (d) (a) (e) Fig.5: (a) Original Image; (b) Masked Image; (c) Threshold Image; (d)After Particle filter; (e) Final OCR. RESULTS:
  • 9. Page | 9 The above code is implemented on still images for working illustration. We have implemented the algorithm on real time videos which shows the practical implementation of our method.
  • 10. Page | 10 3) Ambulance detection: The Ambulance detection is performed in two steps:  Detection of presence of Ambulance in one of the four lanes  Identification of that lane i) Detection of presence of Ambulance in one of the four lanes: The presence of Ambulance is detected using Sound Processing techniques. It involves the analysis of the spectrum of the Fast Fourier Transform(fft) of the sound of the siren of the Ambulance. The following code detected the sound of the siren of the Ambulance:
  • 11. Page | 11  The ambulance siren oscillates between two dominant frequencies, the code exploits this acoustic characteristic of the siren in order to identify it.  First of all the input signal is passed through a band pass filter and signal between 800khz and 1500khz is extracted, because dominant frequencies of siren usually lie in this range. We are not interested in other frequency components.  Then we find 5 nodes from this signal and calculate the mean to find approximately the frequency about which the spectrum is spread.  If mean is <900khz, the given signal is not the siren and no further processing is done.  If mean is >900khz , then the signal is further processed.  The fft of the original signal is obtained and the spectrum is divided into 2 parts: 800khz to 1000khz and 1000khz to 1500khz. The peak is found in each of the two segments.  The frequency at which the peak occurs and the amplitude at that frequency is checked. If they are in the prescribed range in both the segments ,then the signal is the siren and the output led is turned on. RESULTS: i) Identification of that lane: We are using RF modules, working at 434Mhz frequency for the detection of the Ambulance Lane. The driver in the lane will transmit the lane number to the RF receiver, which is connected to myRIO . We have used 8-bit address lines as the encryption mechanism so that only the authorized vehicles can transmit the valid signal. When the driver selects a particular lane the address lines at the transmitter acquires a unique 8-bit value. Data at the receiver is valid , only when address pins at receiver and transmitter are same. So there are 4 different 8-bit values of the address line , each corresponding to one lane. The address lines at the transmitter are achieved through hardware connections.
  • 12. Page | 12 At the receiver, myRIO sequentially assigns the 4 8-bit addresses to the receiver address lines. So the receiver is identified only when the received address lines matches any one of the four assigned address lines. 8 Bit 9V Battery (RF Communication) D0 D1 D2 D0 D1 CONTROL BOARD Data at receiver is valid if address pins of transmitter-receiver match D1 D0 D0 D1 D1 D0 1 D0 A D D R E S S L I N E S Vcc RF TRANSMITTER RF RECIEVER Vcc LANE CONTROL SWITCH ENABLE D0 D1 D2 myRIO O/p 8 pins to led's of 4 lanes 8-Bit Address Lines