Automatic Number Plate Recognition

Automatic Number Plate Recognition Seminar Report
Abstract
Automatic number plate recognition is a mass surveillance method that uses optical
character recognition on images to read the licence plates on vehicles. As of 2006,
systems can scan number plates at around one per second on cars travelling up to 100
mph (160 km/h. !hey can use e"isting closed#circuit television or road#rule enforcement
cameras, or ones specifically designed for the task. !hey are used by various police
forces and as a method of electronic toll collection on pay#per#use roads, and monitoring
traffic activity such as red light adherence in an intersection.
A$%& can be used to store the images captured by the cameras as 'ell as the te"t from
the licence plate, 'ith some configurable to store a photograph of the driver. (ystems
commonly use infrared lighting to allo' the camera to take the picture at any time of
day. A po'erful flash is included in at least one version of the intersection#monitoring
cameras, serving to both illuminate the picture and make the offender a'are of his or her
mistake. A$%& technology tends to be region specific, o'ing to plate variation from
place to place.
'''.seminarsonly.com 1
Introduction
)eing able to identify car number plates *uickly and mechanically is of great benefit to
many businesses and organi+ations for a 'ide variety of applications including security,
crime detection, traffic management and even automatic payment systems (e.g. for car
parks. ,t is only recently that the technology has really come of age, but the systems A-
-ontrols no' install not only provide e"ceptionally reliable results, but also can do so at
acute angles, in any lighting condition and most importantly at high speed.
!he A$%& 'as invented in 1./6 at the %olice (cientific 0evelopment )ranch in the 12.
%rototype systems 'ere 'orking by 1./. and contracts 'ere let to produce industrial
systems, first at 34, 3lectronics then at -omputer &ecognition (ystems (-&( in
5okingham, 12. 3arly trial systems 'ere deployed on the A1 road and at the 0artford
!unnel. !he first arrest due to a detected stolen car 'as made in 1.61.
4assive integration of information technologies into all aspects of modern life caused
demand for processing vehicles as conceptual resources in information systems. )ecause
a standalone information system 'ithout any data has no sense, there 'as also a need to
transform information about vehicles bet'een the reality and information systems. !his
can be achieved by a human agent, or by special intelligent e*uipment 'hich is be able
to recogni+e vehicles by their number plates in a real environment and reflect it into
conceptual resources. )ecause of this, various recognition techni*ues have been
developed and number plate recognition systems are today used in various traffic and
security applications, such as parking, access and border control, or tracking of stolen
cars.
,n parking, number plates are used to calculate duration of the parking. 5hen a vehicle
enters an input gate, number plate is automatically recogni+ed and stored in database.
5hen a vehicle later e"its the parking area through an output gate, number plate is
recogni+ed again and paired 'ith the first#one stored in the database. !he difference in
time is used to calculate the parking fee. Automatic number plate recognition systems
can be used in access control. 7or e"ample, this technology is used in many companies to
grant access only to vehicles of authori+ed personnel.
,n some countries, A$%& systems installed on country borders automatically detect and
monitor border crossings. 3ach vehicle can be registered in a central database and
compared to a black list of stolen vehicles. ,n traffic control, vehicles can be directed to
different lanes for a better congestion control in busy urban communications during the
rush hours.
Mathematical aspects of number plate recognition systems
,n most cases, vehicles are identified by their number plates, 'hich are easily readable
for
humans, but not for machines. 7or machine, a number plate is only a grey picture defined
as a t'o#dimensional function , ( y x f , 'here x and y are spatial coordinates, and f is a
light intensity at that point. )ecause of this, it is necessary to design robust mathematical
machinery, 'hich 'ill be able to e"tract semantics from spatial domain of the captured
image. !hese functions are implemented in so#called 8A$%& systems9, 'here the
acronym 8A$%&9 stands for an 8Automatic $umber %late &ecognition9. A$%& system
means transformation of data bet'een the real environment and information systems.
!he design of A$%& systems is a field of research in artificial intelligence, machine
vision, pattern recognition and neural net'orks. )ecause of this, the main goal of this
thesis is to study algorithmic and mathematical principles of automatic number plate
recognition systems.
:umans define the number plate in a natural language as a small plastic or metal plate
attached to a vehicle for official identification purposes9, but machines do not
understand this definition. )ecause of this, there is a need to find an alternative definition
of the number plate based on descriptors, 'hich 'ill be comprehensible for machines are
segmented using the hori+ontal pro;ection of a pre#processed number plate, but
sometimes these principles can fail, especially if detected number plates are too 'arped
or ske'ed. !hen, more sophisticated segmentation algorithms must be used.
Technology Highlights
!his technology is gaining popularity in security and traffic installations. !he technology
concept assumes that all vehicles already have the identity displayed (the plate< so no
additional transmitter or responder is re*uired to be installed on the car.
!he system uses illumination (such as ,nfra#red and a camera to take the image of the
front or rear of the vehicle, then an image#processing soft'are analy+es the images and
e"tracts the plate information. !his data is used for enforcement, data collection, and (as
in the access#control system featured above can be used to open a gate if the car is
authori+ed or keep a time record on the entry or e"it for automatic payment calculations.
!he =%& system significant advantage is that the system can keep an image record of the
vehicle, 'hich is useful in order to fight crime and fraud (>an image is 'orth a thousand
'ords>. An additional camera can focus on the driver face and save the image for
security reasons. Additionally, this technology does not need any installation per car
(such as in all the other technologies that re*uire a transmitter added on each car or
carried by the driver.
3arly =%& systems suffered from a lo' recognition rate, lo'er than re*uired by practical
systems. !he e"ternal effects (sun and headlights, bad plates, 'ide number of plates
types and the limited level of the recognition soft'are and vision hard'are yielded lo'
*uality systems.
:o'ever, recent improvements in the soft'are and hard'are have made the =%&
systems much more reliable and 'idespread. ?ou can no' find these systems in
numerous installations and the numbers of systems are gro'ing e"ponentially, efficiently
automating more and more tasks in different market segments. ,n many cases the =%&
unit is added as retrofit in addition to e"isting solutions, such as a magnetic card reader
or ticket dispenser/reader, in order to add more functionality to the e"isting facility.
3ven if the recognition is not absolute, the application that depends on the recognition
results can compensate the errors and produce a virtually fla'less system. 7or e"ample,
'hen comparing the recognition of the entry time of a car to the e"it time in order to
establish the parking time, the match (of entry verses e"it can allo' some small degree
of error 'ithout making a mistake. !his intelligent integration can overcome some of the
=%& fla's and yield dependable and fully automatic systems.
Elements of typical A!" systems
A$%& systems normally consist of the follo'ing units@
#amera$s% # that take the images of the car (front or rear side
Illumination # a controlled light that can bright up the plate, and allo'
day and night operation. ,n most cases the illumination is ,nfra#&ed (,& 'hich is
invisible to the driver.
&rame grabber # an interface board bet'een the camera and the %-,
allo's the soft'are to read the image information
#omputer # normally a %- running 5indo's or =inu". ,t runs the =%&
application, 'hich controls the system, reads the images, analy+es and identifies
the plate, and interfaces 'ith other applications and systems.
'oft(are # the application and the recognition package. 1sually the
recognition package is supplied as a 0== (0ynamic =ink =ibrary.
Hard(are # various input/output boards used to interface the e"ternal
'orld (such as control boards and net'orking boards
)atabase # the events are recorded on a local database or transmitted over
the net'ork. !he data includes the recognition results and (optionally the vehicle
or driver#face image file
!he follo'ing illustration sho's a typical configuration of a =%& system (for e"ample,
for 2#lanes#in and 2#lanes#out access control system. !he system (>(ee=ane> is a
typical e"ample of such system. !he (ee=ane application runs as a background 5indo's
application in the %- (sho'n in the center, and interfaces to a set of (ee-ar:ead
camera/illumination units (one for each vehicle 'hich are interfaced by the frame
'''.seminarsonly.com *
grabber. !he application controls the sensors and controls via an ,/A card that is
connected thru a terminal block to the inputs and outputs. !he application displays the
results and can also send them via serial communication and via 003 messages to other
application(s. ,t 'rites the information to local database or to optional remote databases
(via the net'ork.
(ee=ane ma;or features are@
!he program is the basis for a 'ide range of !raffic and Access#control
applications
:andles single or multiple lanes (1#6 lanes, each lane 'ith different
characteristics easily changed by a pass'ord#controlled setting menu
3asy configuration settings@ communication, device interface, sensor types,
capture and illumination modes
'''.seminarsonly.com +
(electable illumination and image capture schemes (to ma"imi+e the results on
different plate *ualities the application s'itches automatically bet'een the B
illumination levels and outputs the result according to built#in intelligence
(ee=ane can run in background (as black bo" and send 003 messages to your
client application (samples and source code are provided
-uts time#to#market # a tailored application can be prepared and integrated in a
fe' days<
(implifies the interface of the (ee-ar recognition package to a &elational
database or a multi#site =%& system (ask for application note
Automatic image recording option (for security checks and debugging
7ull hard'are kit is included (frame grabber, ,A card and interface, sensor
interface, -amera/,llumination unit
&eleased for 5in 2000/C%
'ystem Architecture
(ee=ane is a turn#key system comprises of the follo'ing elements@
o A !# %entium running 5indo's 22/5inC% %ro 'ee#ar ),, # 'hich is used to
analy+e the images and e"tract license plate string.
o #amera-Illumination unit to capture the images ((ee-ar:ead D :i#!ech
(olutionEs =%& camera and illumination unit
o A &rame .rabber$s% # 'hich capture(s the images from the camera units
(handles multiple camera inputs
o I-/ card D input/output board 'ith multiple ,/A discrete lines. !his board
supports the sensors, illumination control and optional gate#open signal. ,t is
connected via a cable to a terminal interface board 'ith easy connections and
indicator lights.
o 'ensors to indicate the presence of the car (a sensor for each lane
o 'ee,ane !he (ee=ane 5indo's application interfaces the hard'are elements
(frame grabber, camera/illumination unit(s, ,A card and sensor. ,t controls the
illumination, reads the video inputs and passes the images to the 0== in order to
obtain the recognition results. !he application displays the image and recognition
results. ,t then e"ports the results using serial communication, messages or disk
files. ,ts man#machine interface supports on#line setting control, 'hich can easily
adapt the application to various types of configurations.
2or3ing /f A!"
!he follo'ing e"ample sho's ho' a typical access#control system 'orks. ,t follo's the
order of the animation above.
7ig 1. !he vehicle approached the secured area, and starts the cycle by stepping over a
magnetic loop detector ('hich is the most popular vehicle sensor. !he loop detector
senses the car and its presence is signaled to the =%& unit.
7ig 2. !he =%& unit activates the illumination (invisible ,nfra#red in most cases and
takes pictures of the front or rear plates from the =%& camera (sho'n at the left side of
the gate. !he images of the vehicle include the plate and the pi"el information is read by
the =%& unitFs image processing hard'are (the frame grabber.
7ig B. !he =%& unit analy+es the image 'ith different image processing soft'are
algorithms, enhences the image, detects the plate position, e"tracts the plate string, and
identifies the fonts using special artificial intelligence methods (such as $eural
$et'orks.4ost =%& units are based on an application running on %- under 5indo's.
Ather systems e"ist that do not re*uire a %- (such as the stand#alone unit sho'n in this
illustration.
7ig G. !he =%& unit checks if the vehicle appears on a predefined list of authori+ed cars,
and if found # it signals to open the gate by activating its relay. !he unit can also s'itch
on a green >go#ahead> light or red >stop> light. !he unit can also display a 5elcome<
message 'ith personali+ed data.
7ig H !he authori+ed vehicle enters into the secured area. After passing the gate its
detector closes the gate. $o' the system 'aits for the ne"t vehicle to approach the
secured area.
Ather types of applications use the information retrieved from the image for different
purposes. 7or e"ample, to prepare a speed or red#light violation ticket. All rely on
automatic image understanding process performed by the =%& unit, 'hich actually
mimics the human mind.
)ifficulties
!here are a number of possible difficulties that the soft'are must be able to cope 'ith.
!hese include@
%oor image resolution, usually because the plate is too far a'ay but
sometimes resulting from the use of a lo'#*uality camera.
)lurry images, particularly motion blur
%oor lighting and lo' contrast due to overe"posure, reflection or shado's
An ob;ect obscuring (part of the plate, *uite often a to' bar, or dirt on the
plate
A different font, popular for vanity plates (some countries do not allo'
such plates, eliminating the problem
-ircumvention techni*ues
3arly A$%& systems 'ere unable to read 'hite or silver lettering on black background,
as permitted on 12 vehicles built prior to 1./B.
5hile some of these problems can be corrected 'ithin the soft'are it is primarily left to
the hard'are side of the system to 'ork out solutions to these difficulties. ,ncreasing the
height of the camera may avoid problems 'ith ob;ects (such as other vehicles obscuring
the plate, but introduces and increases other problems such as the ad;usting for the
increased ske' of the plate.
4any countries no' use licence plates that are retroreflective . !his returns the light
back to the source and thus improves the contrast of the image. ,n some countries, the
characters on the plate are not reflective, giving a high level of contrast 'ith the
reflective background in any lighting conditions. A camera that makes use of infrared
imaging ('ith a normal colour filter over the lens and an infrared light#source ne"t to it
benefits greatly from this as the infrared 'aves are reflected back from the plate. !his is
only possible on dedicated A$%& cameras, ho'ever, and so cameras used for other
purposes must rely more heavily on the soft'are capabilities. 7urther, 'hen a full#colour
image is re*uired as 'ell as use of the A$%&#retrieved details it is necessary to have one
infrared#enabled camera and one normal (colour camera 'orking together.
)lurry images make A-& difficult D A$%& systems should have fast shutter speeds to
avoid motion blur
!o avoid blurring it is ideal to have the shutter speed of a dedicated camera set to
1/1000th of a second. )ecause the car is moving, slo'er speeds could result in an image
'hich is too blurred to read using the A-& soft'are, especially if the camera is much
higher up than the vehicle. ,n slo'#moving traffic, or 'hen the camera is at a lo'er level
and the vehicle is at an angle approaching the camera, the shutter speed does not need to
be so fast. (hutter speeds of 1/H00th of a second can cope 'ith traffic moving up to G0
mph (6G km/h and 1/2H0th of a second up to H mph (6 km/h.
An some cars, to'bars may obscure one or t'o characters of the licence plate. )ikes on
bike racks can also obscure the number plate, though in some countries and ;urisdictions,
such as $e' (outh 5ales, >bike plates> are supposed to be fitted.
(ome small#scale systems allo' for some errors in the licence plate. 5hen used for
giving specific vehicles access to a barriered area the decision may be made to have an
acceptable error rate of one character. !his is because the likelihood of an unauthorised
car having such a similar licence plate is seen as *uite small. :o'ever, this level of
inaccuracy 'ould not be acceptable in most applications of an A$%& system.

Typical applications
=%& applications have a 'ide range of applications, 'hich use the e"tracted plate
number and optional images to create automated solutions for various problems. !hese
include the follo'ing sample applications.
!ar3ing 5
!he plate number is used to automatically enter pre#paid members and calculate parking
fee for non#members (by comparing the e"it and entry times. !he optional driver face
image can be used to prevent car hi;acking.
,n this e"ample, a car is entering a car park in a busy shopping center. !he car plate is
recogni+ed and stored. 5hen the car 'ill later e"it (through the gate on the right side the
car plate 'ill be read again. !he driver 'ill be charged for the duration of the parking. !he
gate 'ill automatically open after payment # or if the vehicle has a monthly permit.
Access #ontrol # a gate automatically opens for authori+ed members in a secured area,
thus replacing or assisting the security guard. !he events are logged on a database and
could be used to search the history of events.
,n this e"ample, the gate has ;ust been automatically raised for the authori+ed vehicle,
after being recogni+ed by the system. A large outdoor display greets the driver. !he event
(result, time and image is logged in the database.
Tolling - the car number is used to calculate the travel fee in a toll#road, or used to
double#check the ticket.
,n this installation, the plate is read 'hen the vehicle enters the toll lane and presents a
pass card. !he information of the vehicle is retrieved from the database and compared
against the pass information. ,n case of fraud the operator is notified.
'''.seminarsonly.com 1*
6order #ontrol # the car number is registered in the entry or e"its to the -ountry, and
used to monitor the border crossings. ,t can short the border crossing turnaround time
and cut short the typical long lines.
!his installation covers the borders of the entire -ountry. 3ach vehicle is registered into
a central database and linked to additional information such as the passport data. !his is
used to track all border crossings.
Traffic control # the vehicles can be directed to different lanes according to their entry
permits (such as in 1niversity comple" pro;ects. !he system effectively reduces traffic
congestions and the number of attendants.
,n this installation the =%& based system classifies the cars on a congested entrance to B
types (authori+ed, kno'n visitors, and unkno'n cars for in*uiry and guides them to the
appropriate lane. !his system reduced the long 'aiting lines and simplified the security
officerEs 'orkload.
'''.seminarsonly.com 1+
Mar3eting Tool # the car plates may be used to compile a list of fre*uent visitors for
marketing purposes, or to build a traffic profile (such as the fre*uency of entry verses the
hour or day.
Tra7el # A number of =%& units are installed in different locations in city routes and the
passing vehicle plate numbers are matched bet'een the points. !he average speed and
travel time bet'een these points can be calculated and presented in order to monitor
municipal traffic loads. Additionally, the average speed may be used to issue a speeding
ticket.
,n this e"ample the car is recogni+ed at t'o points, and the violation sho's the photos of
both locations 'hich 'ere taken on bridges on top of the high'ay. !he average speed of
the car is calculated from both points, and displayed if the speed passed a violation
threshold, and optionally printed.
Airport !ar3ing # ,n order to reduce ticket fraud or mistakes, the =%& unit is used to
capture the plate number and image of the cars. !he information may be used to calculate
the parking time or provide a proof of parking in case of a lost ticket # a typical problem
in airport parking 'hich have relatively long (and e"pensive parking durations.
!his photo sho's the gate of a long#term airport parking. !he car is recogni+ed on entry
and the data is later used to track the real entry time in case of a lost ticket.
#onclusion
Automatic number plate recognition (A$%& is a mass surveillance method that uses
optical character recognition on images to read the licence plates on vehicles. As of 2006
systems can scan number plates at around one per second on cars travelling up to 100
mph (160 km/h. !hey can use e"isting closed#circuit television or road#rule enforcement
cameras, or ones specifically designed for the task. !hey are used by various police
forces and as a method of electronic toll collection on pay#per#use roads, and monitoring
traffic activity such as red light adherence in an intersection.


Automatic Number Plate Recognition

Uploaded by

Copyright:

Available Formats

Automatic Number Plate Recognition

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Automatic Number Plate Recognition

Uploaded by

Copyright:

Available Formats

What are some applications of automatic number plate recognition?

How does ANPR work?

Automatic Number Plate Recognition Seminar Report

You might also like