niyantra 2014

17
0 ADAPTIVE AND AUTONOMOUS TRAFFIC MANAGEMENT SYSTEM (A.A.T.M.S.) USING LabVIEW DELHI TECHNOLOGICAL UNIVERSITY National Conference on Advanced Computing and Research (NCACR)-2015

Upload: nitish-goel

Post on 15-Feb-2017

194 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: niyantra 2014

0

ADAPTIVE AND AUTONOMOUS TRAFFIC MANAGEMENT SYSTEM

(A.A.T.M.S.) USING LabVIEW

DELHI TECHNOLOGICAL UNIVERSITY

National Conference on Advanced Computing and Research (NCACR)-2015

Page 2: niyantra 2014

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.

Page 3: niyantra 2014

OUR OBJECTIVES…

Density/number of vehicles

calculation

License plate Detection &

Identification

Ambulance Detection &

Lane Identification

Page 4: niyantra 2014

SOFTWARE USEDNI LabVIEW 2013, NI Vision Acquisition Express 2013, NI

VDM 2013

HARDWARE USEDNI myRIO, RF Module(434Mhz) Powered USB Hub, Switches

Page 5: niyantra 2014

DENSITY/NUMBER OF VEHICLES CALCULATION

DETECTION OF VEHICLES IN EACH LANE

TRAFFIC LIGHTS CONTROL ALGORITHM

Page 6: niyantra 2014

DETECTION OF VEHICLES IN EACH LANE

Original

Image

Background

Subtraction

Threshold

Morphology

Particle Analysis

RESULTS

Background Model Current Frame Subtracted Image After Morphology

After Particle Analysis

Page 7: niyantra 2014

TRAFFIC LIGHTS CONTROL ALGORITHM

The duration of active green signal depends upon the number of vehicles/vehicle density in each lane.

The loop starts with Lane 1 and iterates till Lane 4 taking one frame of video at a time.

If Ambulance is detected, an interrupt is generated to give active green signal to that lane after which the loop iterates sequentially like before.

An over-ride switch is provided. When this is active, an equal delay is provided to each lane independent of the number of vehicles. This is useful if cameras are to be uninstalled for maintainance work.

The data of number of cars of each lane is stored in a 1-D array for processing.

Page 8: niyantra 2014

ASSUMPTIONS…

At a time, only one lane has an active green signal

Left turn is always free

The duration of forward green signal and right turn green signal is same.

Illustration

Page 9: niyantra 2014

RESULTS IN LabVIEW

LANE 1

LANE 2

LANE 3

LANE 4

Page 10: niyantra 2014

APPLICATIONSPREVENTION OFHEAVY

TRAFFIC JAMSThe system sets different delays for different lanes depending upon the number of vehicles in that lane, thus preventing unnecessary traffic jams.

REAL TIME TRAFFIC UPDATE

The real time traffic information of traffic density can be updated on the traffic website wherein the users can easily access that information.

SOLUTION TO UNNECESSARY

WAITINGOur system provides the method which successfully reduces the waiting time of commuters at traffic junctions even if there is no traffic.

Page 11: niyantra 2014

LICENSE PLATE DETECTION AND IDENTIFICATION

LICENSE PLATE LOCALISATION

SEGMENTATION AND OCR

Page 12: niyantra 2014

LICENSE PLATE LOCALISATIONOrigin

al Image

Color Plane

Extraction

Threshold

Binary Inversion

Morphology

Particle Filter

RESULTS

Original Image

Image Mask

After Binary Inversion

After Morphological Operations

After Image Mask

Page 13: niyantra 2014

SEGMENTATION AND OCROrigin

al Image

Image Mask

Color Plane Extraction Threshold

Binary Inversi

onMorpholo

gyRESULT

S

Image Mask OCR

Original Image

Masked Region

ThresholdImage

Particle Filter

Final OCR

Page 14: niyantra 2014

RESULTS IN LabVIEWVehicle 1Vehicle 2Vehicle 3

The above algorithm is also implemented on Real Time Videos

which shows the practical

implementation of our project

Page 15: niyantra 2014

APPLICATIONSGENERATION OF

DATABASE This system automatically generates the database of identified license plates from which the defaulters can be identified.

AUTOMATED PARKING SYSTEMIn Multilevel Parking Systems,

different spots can be pre-assigned to different users. So the users is identified via Registration Plate detection and their vehicles are parked automatically.

AUTOMATED TOLL COLLECTION

The commuters can be identified via license plate detection and automatically the toll can be debited from their account.

AUTOMATIC GENERATION OF E-

CHALLANThe commuters can be identified via license plate detection and automatically the challan can be debited from their account.

AUTOMATIC TRACKING OF TRAFFIC LIGHT

DEFAULTERS

Page 16: niyantra 2014

FUTURE SCOPE....REAL TIME TRAFFIC

INFORMATIONUPDATE ON TRAFFIC

DEPARTMENT WEBSITEAUTOMATIC GENERATION OF E-CHALLAN

AUTOMATIC TRACKING OF TRAFFIC LAWS DEFAULTERS

IMPLEMENTATION OF ABOVE SYSTEM AT NIGHT USING

THERMAL CAMERAS

DETECTION OF OTHER EMERGENCY VEHICLES

Page 17: niyantra 2014

THANK YOU