ABHINANDAN SINGH - 08070121202 MAYANK KISHORE PATHIKRIT GUHA - 08070121224 - 08070121232

Under the Guidance of

In partial fulfilment of
B.Tech ( E& TC )




This is to certify that the project entitled LICENSE PLATE RECOGNITION SOFTWARE submitted by _ Abhinandan Singh _ , Roll No. _ 08070121202 _, is record of

bonafide work carried out by her/him , under my guidance, in partial fulfillment of the requirement for the award of Degree of Bachelor of Engineering ( Electronics & Telecommunication ) of Symbiosis International University.

Date :

Place :

Prof. Mrudang Shukla
Guide , Department of E & TC Engineering Symbiosis Institute of Technology Pune – 412 115.

Prof. Dr. Narayan Pisharoty
Head, Department of E & TC Engineering Symbiosis Institute of Technology Pune – 412 115.




This is to certify that the project entitled LICENSE PLATE RECOGNITION SOFTWARE submitted by _ Mayank Kishore _ , Roll No. _ 08070121224 _, is record of

bonafide work carried out by her/him , under my guidance, in partial fulfillment of the requirement for the award of Degree of Bachelor of Engineering ( Electronics & Telecommunication ) of Symbiosis International University.

Date :

Place :

Prof. Mrudang Shukla
Guide , Department of E & TC Engineering Symbiosis Institute of Technology Pune – 412 115.

Prof. Dr. Narayan Pisharoty
Head, Department of E & TC Engineering Symbiosis Institute of Technology Pune – 412 115.


_ 08070121232 _. Dr. Mrudang Shukla Guide . in partial fulfillment of the requirement for the award of Degree of Bachelor of Engineering ( Electronics & Telecommunication ) of Symbiosis International University.4 CERTIFICATE This is to certify that the project entitled LICENSE PLATE RECOGNITION SOFTWARE submitted by _ Pathikrit Guha _ . under my guidance. 4 . Roll No. Date : Place : Prof. Department of E & TC Engineering Symbiosis Institute of Technology Pune – 412 115. is record of bonafide work carried out by her/him . Narayan Pisharoty Head. Prof. Department of E & TC Engineering Symbiosis Institute of Technology Pune – 412 115.

08070121224 Pathikrit Guha . ______________________________________ 2.08070121232 Is approved for the degree of BACHELOR OF ENGINEERING – Electronics & Telecommunication Symbiosis International University.08070121202 Mayank Kishore . Pune.5 The Project entitled “LICENSE PLATE RECOGNITION SOFTWARE” By Abhinandan Singh . Examiners: 1. _______________________________________ Date : Place: 5 .

encouragement. enlightenment and cooperation. With deep sense of gratitude we express our sincere thanks to our esteemed and worthy supervisor Mr. Head of the Department of Electronics & telecommunication. Pune who has been a constant source pf inspiration for us throughout this work. We are also thankful to all the staff members of the Department for their full cooperation and help. Narayan Pisharoty. ability and strength in us to complete this work. Pune for his valuable guidance in carrying out this work under his effective supervision. Symbiosis Institute of Technology. We would have never succeeded in completing our task without the cooperation.6 ACKNOWLEDGEMENT The real spirit of achieving a goal is through the way of excellence and austere discipline. Department of Electronics & Telecommunication. Mrudang Shukla. We shall be failing in our duties if we do not express our deep sense of gratitude towards Dr. inspiration ad help provided to us by various personalities. Above all we render our gratitude to the ALMIGHTY who bestowed selfconfidence. Symbiosis Institute of technology. 6 . Assistant Professor.

1 Problem Statement 2.2 Image Processing 4. 10 11 12 14 14 14 15 15 15 17 18 20 22 24 24 25 26 27 28 29 34 34 34 34 34 34 35 37 37 38 . Development History Chapter 8.2 How the Vehicle License Plate System works 3. 8.1 Introduction to matlab 4.1Image Cropping 9.2 Conversion of RGB Images to Binary Images 7 Page no.7 Table of Contents Contents List of Figures Abstract Chapter 1.6 Save to file in given format Chapter 9. Algorithm 8.3 Objectives 3. Literature Survey Chapter 2.5 Number identification 8. Introduction 3.1 Algorithm for ANPR system 8.2 Features of project Chapter 3.3 Comparison Between MATLAB and Artificial Intelligence Chapter 5.3 Convert image into binary. Imaging Hardware Chapter 7.5 Various Types of Vehicle License Plate 3.1 Project Background 3. Image Acquisition 9. Components Chapter 6.4 Segmentation 8.2 Input Image from file. Problem Definition 2.4 Constraints 3.6 Commercial Products Chapter 4 Platform / Technology 4. 8.

2 Discussion Chapter 19 Difficulties and Recommendations 8 40 42 42 42 43 43 44 44 45 45 45 45 46 48 48 49 50 50 50 52 52 53 53 55 57 57 58 58 61 61 63 65 .3 Filtering of Digits 12.4 Identify and remove away the small connected objects Chapter 13 Image Segmentation 13.2 Vehicle Quantization and Equalization 15.1 Imcomplement 10.3 Checking and verification of the Static Chapter 16 List of Functions Used Chapter 17 Code 17.2 Binary Area Open(Bwareaopen) 12.1 Algorithm to extract digits 17.3 Colour Image Processing Chapter 10.2 Algorithm to normalisation of characters 17.8 9.2 Database Chapter 14 Character Recognition 14.2 Global Thresholding 11.1 Experimental results 18. Image Enhancement 10.2 Template Matching Chapter 15 Analysis of Result 15.3 Adaptive thresholding Chapter 12 Image Noise 12.1 Normalization 14.1 Feature Extraction from Digital Image 13.2 Medfilt2 Chapter 11 Thresholding 11.1 Threshold Selection 11.1 Filtering 12.3 Algorithm to read characters Chapter 18 Result Analysis 18.1 Extracting of Individual Digits 15.

9 Chapter 20 Critical Review and Reflection Chapter 21 Conclusion and Future scope Annexure A: Project Planner and Progress Report References 67 68 71 72 9 .

...........2 A car approaching a License Plate Recognition System.............................................39 9...........36 9............1 Original image of the vehicle...1 Overview of the vehicle license plate system...........................3 Indian car license plate....................2 Binary image after filtering.......................................................................................................................................................................20 4.......4 System Block Diagram...........................48 13...........................................................18 3..............................................5 Commercial and Private vehicle license plate..........................25 8.....24 4...............................................................................3 Characteristic function of binary image..............39 9..40 9.......................................41 12..........................................1 Comparison of MATLAb with other softwares........................................................5 Histogram of object and background which is close.......1 Segmented image.............................37 9...................16 3..................................1 Automatic Number Plate Recognition System................1 Captured image of vehicle1................................55 10 ................8 Binary image of the vehicle...........................................54 15.............................50 15...................3Morphologically open binary image...............................................53 15..........2 Block diagram of Image processing system...........37 9.......10 List Of Figures 3...............................1 Binary image with noise..17 3...............................................................38 9...........................................................................47 12....................................................................19 3..................4 Histogram of light object with dark background..................................................41 9........6 RGB schematic block diagram....53 15................7 Captured image of the vehicle.....4 Morphologically open binary image......................2 Cropped image of the vehicle.............................2 Captured image of vehicle2.........

border and customs checkpoints. segmentation of characters and recognition of plate characters. a smart and simple algorithm is presented for vehicle‘s license plate recognition system. This system also helps to conduct the graphic images of the vehicles which can be further stored in the database in text format reducing size of data to be stored. The performance of the proposed algorithm has been tested on real images.  Implementing a file management system or database for storing the images of vehicle license plate. etc. For extracting the plate region.).11 ABSTRACT Automatic Vehicle Identification (AVI) has many applications in traffic systems (highway electronic toll collection. Based on the experimental results. And finally statistical based template matching is used for recognition of plate characters. numbers and characters Keywords     Character recognizer license plate recognition Segmentation Cropping 11     template matching Median Filter Connected Components Masking. License Plate Recognition is an effective form of AVI systems. processing them and using processed data for further processes like storing. This system is based on the image processing system. we noted that our algorithm shows superior performance in car license plate recognition.  Plate localization and extraction: To obtain the vehicle plate sub image. The neural network will be trained off-line with the characters and numbers. red light violation enforcement. allowing vehicle to pass or to reject vehicle. We used neural network for recognition of vehicle license late character. . The proposed algorithm consists of three major parts: Extraction of plate region. edge detection algorithms and smearing algorithms are used. In this study. The general algorithm involves the following steps:  Image Processing : The image captured is pre-processed and reduction in the contrast. In segmentation part.  Character Segmentation/Recognition: Resample and threshold in order to isolate the license plate and vehicle license plate character. This system helps in the functions like detection of the number plates of the vehicles. filtering and some algorithms are used.  Evaluating the performance of the algorithm and compare the performance with other reported work.

Eventually. such as Hough transform. Due to these features. The input image being first processed to enrich and enhance boundary line-information by using such algorithms as the gradient filter. To add the further mile for this step we have presented a License plate recognition software . to detect lines. In the developed western countries many big factories. The early approaches were based on characteristics of boundary lines. but distinguishes each as unique. etc. Vehicle License Plate Recognition is an image processing system whereby it is used to recognize the vehicles by identifying the license plate. License plate identification/recognition (LPI/R) is one form of ITS technology that not only recognizes and counts vehicles. reliability and versatile nature of the computer. and resulting in an image formed of edges.12 1 LITERATURE SURVEY It is almost known that the world is nowadays being more dependent in the computer than the human manpower. contrast. communication centers all of them are under the control of the computer because that‘s not the place where human hand can reach up. In this approach. This approach was commonly used in finding text in images. The image thus processed was converted to its binary counterpart and then processed by certain algorithms. text regions were discovered using statistical properties of text like the variance of gray level. and could well be used for discovering and designating candidate number plate areas as they include alphabets and numerals. It is basically use for traffic and security purposes. All of the systems discussed above have some kind of limitations for example they are plate size dependent. this method could be used to detect the similar properties in a certain image and locate the position of number plate regions. work only in certain conditions or environment like indoor images etc.The project is based on image processing technology using popular software MATLAB. symmetry. This approach focuses on some salient properties of vehicle plate images such as their brightness. couples of 2-parallel lines were considered as a plate-designate. such as electronic toll collection and red-light violation 12 . For some applications. color dependent. The problem of automatic VNP recognition is being studied since the 90‘s. etc. edge densities in the region. Another approach was based on the morphology of objects in an image. nuclear plants. Today in the developed countries more than 90% of the job is done by automated system and this is due to the accuracy. The third approach was based on statistical properties of text. angles. number of edges.

For implementing LPR system we have used digital image processing technique. crossing red light. occupying lanes reserved for public transport. like commercial vehicle operations or secure-access control. LPI/R records a license plates alphanumeric so the vehicle owner can be assessed the appropriate toll or fine. etc. a vehicle's license plate is checked against a database of acceptable ones to determine whether a truck can bypass a weigh station or a car can enter a gated community or parking lot. breaking speed limits . The LPR system can be used to traffic control management for recognize vehicles that commit traffic violation. such as entering restricted area without permission . License plate recognition (LPR) is a new tool for automatic vehicle and traffic monitoring by using digital image processing.13 enforcement. 13 . In others.

They are used by various police forces and as a method of electronic toll collection on pay-per-use roads and cataloging the movements of traffic or individuals. owing to plate variation from place to place. ANPR can be used to store the images captured by the cameras as well as the text from the license plate. Previously. So this project is developing to replace human to monitor the car and automatically capture the image. with some configurable to store a photograph of the driver. human is needed to observe and list the user car plate number manually. the system can automatically display the status of the car which it will compare between the car plate numbers recognized with the database. high error rates. Concerns about these systems have centered on privacy fears of government tracking citizens' movements. Besides that. and increased government spending. 14 .1 Problem statement: Automatic car license plate recognition (CLPR) system is implemented to help the human to automatically detect plate number without human supervision. They can use existing closed-circuit television or road-rule enforcement cameras.2 Features Automatic number plate recognition (ANPR) is a mass surveillance method that uses optical character recognition on images to read vehicle registration plates. 2. or ones specifically designed for the task.[ANPR technology tends to be region-specific. misidentification.14 2 Problem Definition 2. Systems commonly use infrared lighting to allow the camera to take the picture at any time of the day.

there was also a need to transform information about vehicles between the reality and information systems. In entrance gate. number plates are used to identify the vehicles. Automatic Line Tracking Robot (ALR) is used in this project as a vehicle which contains circuitry for moving in a guided track. Vehicle License Plate Recognition is an image processing system whereby it is used to recognize the vehicles by identifying the license plate. Automatic number plate recognition systems can be used in access control. In traffic control. ANPR systems installed on country borders automatically detect and monitor border crossings. The cycle will start when the vehicle steps over the detector. such as parking. or tracking of stolen cars. the vehicle will stop at the car gantry. It will have mechanism to detect the opened and closed door. 15 .1 Project Background Massive integration of information technologies into all aspects of modern life caused demand for processing vehicles as conceptual resources in information systems because a standalone information system without any data has no sense. For example. It will activate a signal to the Vehicle License Plate System of the presence of the vehicle. When a vehicle later exits the place through the gate. This can be achieved by a human agent. or by special intelligent equipment which is be able to recognize vehicles by their number plates in a real environment and reflect it into conceptual resources. access and border control. Each vehicle can be registered in a central database and compared to a black list of stolen vehicles. . number plate is recognized again and paired with the first-one stored in the database and it is taken a count.15 3 INTRODUCTION 3. vehicles can be directed to different lanes for a better congestion control in busy urban communications during the rush hours. 3. various recognition techniques have been developed and number plate recognition systems are today used in various traffic and security applications. number plate is automatically recognized and stored in database and black-listed number is not given permission. Because of this. When a vehicle enters an input gate. In some countries.2 How the Vehicle License Plate System works Firstly. It also will have capacity to park in the given parking area. this technology is used in many companies to grant access only to vehicles of authorized personnel.

Lastly. Next. the characters will be recognized by using Neural Network. the images will be enhance. we also need to look into other facts of variables that are not constant. the gantry will open and allowed the vehicle to pass through. the system will try to match the recognized vehicle plate number with the car plate database. If ―Access Granted‖. The system will read the information pixels of the vehicle and run the recognition process. Diagram is illustrate in Fig 1. Camera can also focus on the face of the driver and save it for security reason. It is essential and important to determine the facts which will able to influence the operations and recognition proficiency. locating the vehicle plate position and extract the characters from the vehicle plate.1 Besides. Thirdly. illumination (infra-red) will be activated and images of the front picture of the vehicle will be taken. Fig 3. Below are the non-constant variables which will affect the accuracy of recognition:  Speed of the vehicle 16 . Next. the Vehicle License Plate Recognition also provide an advantage by keeping the image of the vehicle in which it will be useful for crime fighting. There are difficulties for Vehicle License Plate Recognition in which it will affect the efficiency and accuracy of the system. Besides analyzing. the system will apply certain algorithm to analyses the vehicle image.16 Secondly.

Fig 3. In the past. the characters will be recognized and displayed on the Graphical User Interface. The software could lead to a cheaper and faster way of enhancing and determined the performance of the recognition system.3 Objectives The overall objective of the project is to develop a system to recognize vehicle license plate from a car at a gate entrance of a parking lot.2 A car approaching a License Plate Recognition System 3. we would need to improve or rewrite the algorithm to improve the accuracy. Bent type) Vehicle license plate orientation Type of vehicle fonts character. Besides. there has been similar project implemented but had poor accuracy. the system can also serve as a security purpose whereby it can spot on any wanted or stolen vehicles. The system will be based on a Personal Computer such that it will generate report on the vehicle license plate it has captured. I 17 .17       Weather condition Type of Vehicle Distance between vehicle license plate and the camera Type of plate ( Rectangular. Thus. Once the vehicle license plate is captured.

Try on zoom in image of the car and image consists of headlamp.3 18 . it is not advisable for me to cover all aspects in this project. Deal with only Indian Car License Plate with Times new roman font. Thus. Captured image of vehicle at fixed angle. it is not advisable to include all of the possible cases. we managed to come up with the progress guideline as to what need to be included in this project. However. Captured images on location where light is proportional. The image should be taken with the height of 50cm to 70 cm above the ground level. due to the limited time frame given. There will be no motion capture image. Fig 3. Thus. Captured image of vehicle at fixed distance. There is definitely a lot more room for further improvement on this project. 3.18 will address a set of constraints and focus on the design of the algorithm to extract the vehicle license plate in order to improve the accuracy. we have to set a list of constraints to make the project more systematic and manageable. The constraint is listed as below:           Image taken only when vehicle is stationary.4 Constraints Due to limited time that we possess and dealing with image vision software. The vehicle license plate position should be captured centred. Take only the front view image of the car. in discussion with my tutor.

4 19 .19 System block Diagram Fig 3.

5 Various Types of Vehicle License Plate  What is Car license Plate Car license Plate is a rectangular metal plate which consists of a combination sequence of alphabet characters and numbers issued by the government for identification purpose of registered vehicle. A letter(s) is prefixed when the 4 digit number runs out and then two letters and so on. Fig 3. DL 2C 0001).g. The third part is a 4 digit number unique to each plate. Commercial vehicles such as taxis and trucks have a yellow background and black text (e..  Current Format The current format of the registration index consists of 3 parts    The first two letters indicate the state to which the vehicle is registered.5 20 . Due to heavy volume of vehicle registration. witnesses usually remember the initial area code letters — it is then quite simple to narrow down suspect vehicles to a much smaller number by checking the database without having to know the full number.20 3. Vehicles belonging to foreign consulates have white lettering on a light blue background (e. Instead they have the Emblem of India in gold embossed on a red plate. 22 UN 14). This scheme of numbering has some advantages:   the State or District of registration of a particular vehicle In the case of a police investigation of an accident or vehicle-related crime.  Format Vehicle owner who register the car as private vehicle have a white background and black text. The next two digit numbers are the sequential number of a district.g. the numbers were given to the RTO offices of registration as well. The President of India and state governors travel in official cars without licence plates.

For public utility vehicles like buses. 21 . This mechanism is used for unique identification. To register a vehicle. Other numbering Other numbering includes the special numbers allotted to public transportation such as buses. For example. confirm that the engine and chassis numbers are identical to what is written in the application and issues a permanent registration certificate which is usually valid for 20 years. the others being a valid insurance certificate. this prevents the number being read wrongly in case the plate (and/or the vehicle bearing it) is upside down. The permanent registration certificate is one of the four important documents a vehicle plying on the road should always have. where a Motor Vehicle Inspector will verify the applicant's address and other details. it has to be presented to the RTO's office. trucks. taxis and pick-up vans. which stand for United Nations. Diplomatic plates Vehicle belonging to foreign missions bear the plates UN. CD or CC. probably a country close to the letter A or B. This is valid for one month. The next two digits (or the two preceding the arrow) signify the year in which the Military procured the vehicle. The first (or the third) character is always an arrow pointing upwards. followed by the serial number. a pollution under control (PUC) certificate and the driver's licence. there are a number of additional documents like a roadworthiness certificate and a transportation permit. a vehicle bearing the number 77 CD xxxx in India refers to a vehicle owned by either the United States mission in India or by a person working with the mission. taxis and auto-rickshaws. Diplomatic Corps or Consular Corps respectively. Temporary numbers As soon as a vehicle is purchased. The letter ending after the serial number indicates the class of the vehicle. The next is the base code.21 Military vehicles Military vehicles have a unique numbering system unlike any other licence numbers. A diplomatic plate numbered 13 CC xxxx would refer to country 13. the dealer of the vehicle issues a temporary licence sticker known colloquially as a TR (To Register) number. The numbers are registered by the organization which run the services and are usually printed on the side of the vehicle. during which the owner must register the vehicle in the controlling RTO of the area in which the owner is residing to get a standard licence plate. The numbers are registered by the Ministry of Defence in New Delhi.

Current LPR system read Latin (A-Z) and Korean (Hangul) letter and Arabic number (09). it automatically locates vehicle license plates and reads the numbers accurately each time every time.1 IMPS (Integrated Multi-Pass System) An IMP [26] is a Singaporean commercially developed license plate recognition system.6 Commercial Products The various products in the market today are described briefly below. however. It is a high performing robust system that gives consistent results under all weather conditions.22 3.6.6. With milliseconds the LPR system locates. 7 days a week. • Reads accurately in most weather conditions. the LPR can be programmed to read any language or symbol in any alphanumeric combination or context on both retro and non-retro reflective plates. VISPA provides for state-of-the-art video technology. Using advanced image processing and artificial intelligent techniques such as AI best first breadth-wise search algorithm. Features of this LPR technology includes. 3. easy installation and has accessories and features for most parking security surveillance needs. combined template and neural network recognizers. The system's reliability and flexibility allow it to accommodate some of the most stringent needs in some of the worst conditions. 3. 3. • Reads accurately at highway speeds. captures and identifies a vehicle's license plate data and makes a read decision. • Works 24 hours a day. fuzzy logic and an arsenal of image processing tools. uses video imaging for better recognition.2 Perceptics Perceptics [27] is the world leader in license plate reader technology. 22 . • Automatic and within milliseconds. identification and improved security.3 Vehicle Identification System for Parking Areas (VISPA) PPI's Vehicle Identification System for Parking Areas (VISPA) [28].6.

• Compatible with standard hardware and software.Detects and reads Vehicle license plates for parking.An image of the car and/or the driver (depending on the location of your camera) will be taken as soon as the car approaches the triggering device. Available as a complete system which is based on a background Windows application. SeeContainer Identification System:. The recognition result is then logged together with the images. The image will be linked to the ticket. The VISPA controller with an integrated frame grabber card for 4. Windows DLL or Linux library. 23 . • Can be customized according to specific user needs. This is the main advantage of vision based recognition.crane mounted Container recognition system. VISPA is available in two forms Basic Version: . The license plate can then be encoded on the ticket. or 16 cameras automatically identifies the license plate from the video image and stores it in a database.23 Features are • Open architecture to most common video-systems. law enforcement and security applications.Tracks and reads Shipping container identification marking. There product includes.6. as a stand-alone turn-key version. Available as complete systems. and transmits the ID string to the port or gate computer. traffic surveillance. or SeeCrane . access control. 8. the system read the camera images and extract the identification data from the images. or to a client process. Enhanced Version:. 3. The basic system version connects to 4 cameras and can be upgraded to 8 cameras. such as SeeGate . or in form of different special-task systems. SeeCar License Plate Recognition:. There technology is based on computer vision.4 Hi-Tech Solution Hi-Tech Solutions [29] is a system and software company that develops cutting edge optical character recognition (OCR) solutions by implementing the company's unique image processing software and hardware in a wide range of security and transportation applications.License Plate Identification. the records include both the image plus the extracted result.a recognition system for the Tracks and Containers.

Besides. Comparison of MATLAB with other program is shown in the below diagram. 24 . The table below show images/graphs format that is supported by ―imread‖ with MATLAB. The task is to construct the algorithm and recognizes can be done using MATLAB. Fig 4. The advantages of MATLAB can be classified as follows: Platform independence.24 4 Platform/Technology 4. predefined function and deviceindependent plotting.1 INTRODUCTION TO MATLAB MATLAB is chosen as the main development for the license plate recognition application. MATLAB is also very efficient as it have built-in-function tools for neural network and image processing.1 MATLAB read images using the function ―imread‖. Graphical User interface and MATLAB compiler.

Traffic Control etc.tiff . Face Detection.png .25 Format Name TIFF JPEG GIF BMP PNG XWD Description Tagged image file format Joint photographs Expert Group.Jpeg .gif .bmp . This useful information may be the dimensions of an engineering component. Engineering.2 Image Prcessing Image Processing generally involves extraction of useful information from an image. . size of diagnosed tumour. Quality Control. or even a 3D view of an unborn baby.2 25 .Jpg . Graphic Interchange Format Windows bitmap Portable network Graphics X Window Dump Extension . The main areas of application of Image Processing are Bio-Medical.xwd 4. .tif. Block Diagram of Image Processing Steps Fig 4.

3. 3.  Advantage of MATLAB 1. Not good in common sense reasoning. MATLAB possess power graphic visualization tools. 3. MATLAB coding is shorter and simpler. Non-programmers can easily understand the rules of artificial intelligence. 2. Good for solving complex problems where humans or programmer have expert knowledge in it.26 4. We can changed and add individual rules. . There are difficulty in getting expert knowledge and putting it in used in the system. Algorithm used in MATLAB can be converted to use in hardware. Expensive. 4. 26 . 5. The system will perform faster when using “Matrix Approach”. 5. 4. Slow response.  Disadvantage of Artificial Intelligence: 1. Unpredictable. 2. 2.3 Comparison Between MATLAB and Artificial Intelligence  Advantage of Artificial Intelligence: 1. Data can be represented in terms of matrices or vectors.

normalize and enhance the image of the number plate. there is a requirement to forward images to the remote server. Often in such systems. date-time. now small data packets. ANPR systems are generally deployed in one of two basic approaches: one allows for the entire process to be performed at the lane location in real-time. can easily be transmitted to some remote computer for further processing if necessary. such as those found in the London congestion charge project. the information captured of the plate alphanumeric. and the other transmits all the images from many lanes to a remote computer location and performs the OCR process there at some later point in time. and any other information that is required is completed in somewhere around 250 milliseconds. lane identification. or stored at the lane for later retrieval. and then optical character recognition (OCR) to extract the alphanumeric of the license plate.27 5 COMPONENTS The software aspect of the system runs on standard home computer hardware and can be linked to other applications or databases. there are typically large numbers of PCs used in a server farm to handle high workloads. 27 . In the other arrangement. When done at the lane site. and this can require larger bandwidth transmission media. It first uses a series of image manipulation techniques to detect. This information.

To avoid blurring it is ideal to have the shutter speed of a dedicated camera set to 1/1000 of a second. In slow-moving traffic. installers should carefully consider the positioning of the camera relative to the target capture area. In some countries. and so cameras used for other purposes must rely more heavily on the software capabilities.28 6 IMAGING HARDWARE At the front end of any ANPR system is the imaging hardware which captures the image of the license plates. slower shutter speeds could result in an image which is too blurred to read using the OCR software. giving a high level of contrast with the reflective background in any lighting conditions. License plate capture is typically performed by specialized cameras designed specifically for the task. Factors which pose difficulty for license plate imaging cameras include speed of the vehicles being recorded. varying ambient lighting conditions. or when the camera is at a lower level and the vehicle is at an angle approaching the camera. 28 . Shutter speeds of 1/500 of a second can cope with traffic moving up to 40 mph (64 km/h) and 1/250 of a second up to 5 mph (8 km/h). This returns the light back to the source and thus improves the contrast of the image. the characters on the plate are not reflective. Exceeding threshold angles of incidence between camera lens and license plate will greatly reduce the probability of obtaining usable images due to distortion. Many countries now use license plates that are retro reflective. A camera that makes use of active infrared imaging (with a normal colour filter over the lens and an infrared illuminator next to it) benefits greatly from this as the infrared waves are reflected back from the plate. the shutter speed does not need to be so fast. To maximize the chances of effective license plate capture. Further. headlight glare and harsh environmental conditions. however. Most dedicated license plate capture cameras will incorporate infrared illumination in order to solve the problems of lighting and plate reflectivity. This is only possible on dedicated ANPR cameras. when a full-colour image is required as well as use of the ANPR-retrieved details it is necessary to have one infrared-enabled camera and one normal (colour) camera working together. Since the car is moving. License plate capture cameras can now produce usable images from vehicles travelling at 120 mph (190 km/h). especially if the camera is much higher up than the vehicle. Manufacturers have developed tools to help eliminate errors from the physical installation of license plate capture cameras.

the patrol officer is provided with information regarding the vehicle. Almost a year later on April 24. 2006). is stored at the National ANPR1 This term is synonymous with ALPR and will be used interchangeably 3 Data Center (NADC) in London. 1992. 2006). and the GPS location of the camera is created. 2008). On April 10. 29 . The central ANPR system. or if it was uninsured (Pughe.29 7 Development History The United Kingdom: In 1992.e. This resulted in an increase in arrests to 100 per year. As of 2001. if it had been involved in another crime. partially destroying the façade of the Exchange‘s offices and extensively damaging the rest of the building (Baltic exchange. In 2002 to 2003. a British company operating a premier global marketplace for shipbrokers and charterers. the time and date of the scan. killed one person. 2003). the databases. i. 2006). 2008). The bombing was massive in size and destroyed several buildings with many others suffering shattered windows. the heart of London‘s financial district (1993 Bishops gate bombing. such as whether the car was stolen. The results of this initial study indicated that officer productivity increased primarily due to the officers not having to spend as much time waiting for hits after they typed in a plate. In addition. the Irish Republican Army bombed the Baltic Exchange. As a part of its intelligence network. stationary (roadside) and mobile (mounted on police cars). the Irish Republican Army detonated a truck bomb at Bishops gate. The blast caused approximately £1 billion in damage. ALPR technology was developed at Cambridge University in the United Kingdom in response to terrorism. and injured 44 others. a JPEG image of the plate is produced. 2006). The United Kingdom is at the forefront in the use of plate recognition technology. NADC stores plate data and lists of suspect vehicles (Pughe. 2006). 10 times the national average (PA Consulting Group. scanning up to 5 million plates a day (Pughe. all 43 police forces in England and Wales were provided with vans equipped with automated number plate technology (ANPR1): ANPR cameras were mounted in vans. officers were able to spend more of their time investigating hits. Instead. When a plate is photographed by an ANPR camera. 1993. a text file containing information on the car registration number. and police forces were provided with computers that were able to store ANPR information in real time (Pughe. the United Kingdom evaluated the use of ANPR with nine police forces. there are 3. as well as a video image of the plate and a video of the vehicle occupants (Pughe. Approximately four seconds after a plate is read by the camera.000 cameras across the United Kingdom.

seven instances of drug seizures (equivalent to approximately 3. the Hampshire Constabulary uses ANPR to prevent and detect terrorism. again suggesting a need for policy outlining a response priority scheme. on average. In nearly two-thirds (61 per cent) of these stops. In total.30 In total. 2003).8% of arrests were made for theft or burglary offences. This initial study also provided the Home Office with descriptions of the quantity of vehicles on the road violating insurance or other traffic regulations. The results of this study were similar to the first study. A second evaluation was conducted between June 2003 and June 2004 with 24 police forces (PA Consulting Group.g. On an annual basis. For instance. three instances involving the recovery of other stolen goods (equivalent to approximately 23. and the number of vehicles owned or operated by persons of interest (PA Consulting Group. or prohibited drivers. and five instances of recovery of other stolen property (PA Consulting Group. an average. 10. In effect. workload concerns continued to exist. 30 . 2004). However. In effect. One-quarter of the arrests were for disqualified. uninsured. ANPR technology continues to be used across the United Kingdom. over 1.3% for theft of or from a vehicle.300 pounds total). 2007).543 vehicles were stopped by intercept teams resulting in 13. however. the number of vehicles used in the commission of other criminal offences. In 2005 and 2006. approximately one out of every 200 cars photographed by the ANPR cameras was stopped by an ANPR intercept team. The Metropolitan Police Service operates four ANPR units who work across London. volume crime. Over this one year period. The results of this evaluation also demonstrated that the ANPR technology contributed to a substantial increase in arrests (Pughe. and 8. A further 16. 200 arrests per month (Metropolitan Police Service. and fatal and serious injury road traffic accidents (Hampshire Constabulary. equivalent to approximately one stop per hour. arrests were made in nearly 4-8% of the vehicle stops. two seizures of weapons and/or firearms. 2006). 2003). Currently. no date). the data also highlighted the limited ability of the intercept teams to respond to the volume of hits they received. 180. the intercept team took some action (e. The results also indicated that the police could expect a substantial increase in the amount and value of goods they recovered. There are deployment teams who are used to assist in targeting hot spots for vehicle and other crime. 2003).2% for drug offences. police were only able to respond to 13% of hits (PA Consulting Group.000 pounds).499 arrests. a constable using ANPR technology could expect to encounter: the recovery of 11 stolen vehicles (equivalent to approximately 68. serious crime. 2004).000 pounds total). enforced an arrest). these units made.000 stolen vehicles were recovered over this period (PA Consulting. ANPR was beneficial to police forces by providing better efficiency in the deployment of officers.

in several jurisdictions. A second data centre provides a backup location for image data. The video streams are transmitted to a data centre located in central London where the ANPR software deduces the registration plate of the vehicle.m. There are currently 1. This list is then compared with a list of cars whose owners/operators have paid to enter the zone – those that have not paid are fined. 150 American cities utilized this technology to catch those who drove through red lights. The registered owner of such a vehicle is looked up in a database provided by the DVLA. The United States Over the past several years. Fines for traveling within the zone without paying the charge are £60 per infraction if paid before the deadline. Both front and back number plates are being captured. and 6:00 p.. ALPR has been used to track potential terrorists.m. primarily in the form of ―red light cameras‖ (Jenkins. Monday to Friday. More recently. There are also a number of mobile camera units which may be deployed anywhere in the zone. The US has also used plate recognition technology at border crossings with Mexico and Canada to track auto theft (Canadian Press NewsWire. 2007). which use Automatic Number Plate Recognition (ANPR) technology. on vehicles going both in and out – this gives up to four chances to capture the number plates of a vehicle entering and exiting the zone. doubling to £120 per infraction thereafter. The London congestion charge is an example of a system that charges motorists entering a payment area. A reduced fee of £9 is paid by vehicle owners who sign up for the automatic deduction scheme. This technology will also be used. In 2007. A recent study of ALPR technology in the state of Ohio emphasized 31 . to catch drivers who violate speed limits (Jenkins. 2006). It is estimated that around 98% of vehicles moving within the zone are caught on camera.500 cameras. leave or move around within the congestion charge zone between 7 a. the United States has also introduced plate recognition technology. These offenders have a photograph taken of their plate by the ALPR cameras and receive a ticket in the mail. 2007).31 Charge zones – the London congestion charge The London congestion charge scheme uses 230 cameras and ANPR to help monitor vehicles in the charging zone. Transport for London (TfL) uses ANPR systems and charges motorists a daily fee of £10 paid before 10pm if they enter.

are prohibited from driving. Still. including reductions in criminal activity. police can use ALPR technology linked to criminal justice and insurance databases to identify those drivers who. whether in the form of additional police on the roads utilizing this technology or the hiring of additional staff to assist in confirming hits. ALPR has the potential to allow Canadian police forces to engage in proactive policing. and maintaining ALPR technology. The state of Ohio studied the effectiveness of ALPR technology over a four month period in 2004. 2004). uninsured. These results were 50% higher than results during the same time period one year previous. As previously discussed. for instance. The cost of warehousing this data. Over the four month period. Nearly one-third (30 per cent) of the tolls are tracked through licence plate recognition technology allowing for faster movement of traffic as the need to stop and physically pay the toll is eliminated (Commission for Integrated Transport. By driving through traffic corridors in cities across Canada. beginning in August 2004. therefore. It allows drivers to use the express route for a toll fee that is recorded electronically. ALPR technology has been used in Toronto for toll-collection purposes. updating the databases. One way to offset these resource increases may be to. recovery of stolen goods. Schuurman (2007) recently produced an analysis of the use of this technology in parking lots in Surrey. must be weighed against the costs recovered through the use of this system. use civilian volunteers. there are several issues that must be considered prior to the widespread use of ALPR technology in Canada. have outstanding warrants. British Columbia. 2004).000 in federal funding.32 the recovery of two stolen trailers demonstrating the technology‘s potential utility for homeland security (McClellan. Additional staffing costs may also be incurred. or are driving uninsured vehicles. criminal justice. the technology also offers police a potentially useful tool for identifying and recovering stolen vehicles. and other agencies that is linked with the ALPR technology in police cruisers. or prohibited drivers. no date). did not provide information on unlicenced. Using information obtained through 32 . Canada ALPR technology in Canada has only recently been implemented for use with traffic and other criminal violations. and/or conducting further searches of licence plates. 2007). and increases in positive public perceptions of the police (Schuurman.000 (McClellan. installing. using $61. wherever possible. The Highway 407 Express Toll Route was introduced in 1997. in addition to the cost of buying. It is important to note that the ALPR technology used in this study was only linked to databases on stolen vehicles and wanted persons and. ALPR was introduced in British Columbia as part of a pilot study in 2006. the state highway patrol apprehended 23 criminal suspects and recovered 24 stolen vehicles valued at US $220. The United Kingdom established a central data warehouse in London containing information from insurance.

one unmarked patrol vehicle drove through 31 Surrey parking lots scanning the licence plates of parked cars over the period of one week. and stolen cars (1. Given that there is limited information available to date on the use of ALPR technology.8 per cent) of hits were for unlicenced drivers. 2007).7 per cent) of the scans accurately captured the car‘s licence plate. Such research can be used to inform deployment strategies. To collect the data for this initial study. in determining whether it is more beneficial to have police equipped with ALPR technology drive their usual patrol routes or travel exclusively within known ―hot spots‖ (Schuurman.4 per cent). prohibited drivers. The remaining hits were for unlicenced (i. prohibited drivers (4. The Hot Lists were for unlicenced vehicles.876 licence plates were scanned. It was concluded that using the technology in a stationary format along major intersections or installing it in mobile units that patrolled the jurisdiction‘s main roads would likely be more effective for identifying stolen vehicles. Canada. research must be conducted to determine the extent to which such technology is useful to Canadian police forces. 33 .9 per cent). four hits every hour (Schuurman. there were. 21.33 the Canadian Police Information Centre (CPIC) and the Motor Vehicle Branch (MVB). Scanned plate images were compared to the CPIC and MVB databases. on average.e. Schuurman (2007) concluded that the deployment of ALPR technology exclusively to parking lots would not be the most efficient use of the system. for example. uninsured drivers. The current study intends to provide some initial data regarding the use of ALPR technology in British Columbia. Schuurman‘s (2007) results indicated that 1.6% of scanned plates resulted in a ‗hit‘. In fact. uninsured) vehicles (23.8 per cent). nearly all (97. especially to locate and recover stolen vehicles. and over two-thirds (69. lists of vehicles of interest2 were populated on a daily basis and uploaded to the patrol car‘s onboard computer. Nearly three-quarters (72 per cent) of the hits occurred between 3pm and 7pm. Of these. Over the course of this study. and stolen vehicles. 2007).


8 Algorithm
8.1 Algorithm for ANPR system 1. Input image from webcam. 2. Convert image into binary. 3. Detect number plate area. 4. Segmentation. 5. Number identification. 6. Save to file in given format. 8.2 Input Image from file. 1. Capture image from webcam. 2. Store the captured image into a image file for further processing. 8.3 Convert image into binary. 1. Identify the intensity of the image. If image intensity = high Reduce intensity Else if intensity = low Increase intensity Else No change. 2. Convert image into grayscale. 3. Calculate appropriate threshold value for the image. 4. Convert the image into binary image using the calculated threshold. 8.4 Segmentation 1. Crop the image. 2. Filter the noise level present in the image. 3. Clip the plate area in such a way that only numbers of plate area extracted. 4. Separate each character from the plate. 8.5 Number identification 1. Create the template file from the stored template images. 2. Resize image obtained from segmentation to the size of template.


3. Compare each character with the templates. 4. Store the best matched character. 8.6 Save to file in given format 1. Open a text file in write mode. 2. Store the character obtained from the number identification process to text file in given format. 3. Close the file.



Overview of the Vehicle license plate System

Fig 8.1

vehicle images will be taken with a Panasonic FX 30 digital camera. this cropping process will highly increase the speed of image processing. we might proceed in using the Matlab function to convert the vehicle JPEG image into gray scale format.1 Image Cropping Image cropping is a recognition process whereby it will extract the smallest rectangle which will contains the edge of the license plate and license plate itself.37 9 Image Acquisition The initial phase of image processing for Vehicle License Plate Recognition is to obtain images of vehicles. For this project. The images will be stored as colour JPEG format on the camera. As the license plate surrounding is of no importance.1: Original Image Fig 9. Next.2 : Cropped Figure 37 . The figure below shows the copping of image: Fig 9. webcam etc can be used to capture the acquired images. 9. Electronic devices such as optical (digital/video) camera.

Binary images is the simplest process and has apply to many other application. The application is as:  Text interpreting  Identify the object orientations Next. It is useful as the information we need can be obtained from the silhouette of the object. Binary Images are obtained by converting the input image into grayscale format. The characteristic function for the image object is classify below: Fig 9. The image is made up of a matrix squares which is called Pixel. The pixel of grey level above the threshold will be set to 1 (equal to 255.1 What is Binary Images It is an image which quantatised into two values representing 0 and 1 or in pixel values of 0 and 255 representing the colour black and white. The conversion is done by using the toolbox in MATLAB (im2bw). then by converting the grayscale image to binary image by thresholding. white) and the rest will be set to 0 (black). the image obtain is being converted to Binary Image.2. Each pixel in the image has a brightness value which is known as grey level. We will obtain white object with black background or vice versa.38 9.3 38 .2 Conversion of RGB Images to Binary Images In order facilitate the next process smoothly and reducing the processing time. 9.

Frequency Fig 9.39 Below will illustrate different type of histogram images Fig 9.5 : Histogram of object and background which is close 39 .4: Histogram of light object with dark background.

40 In this circumstance. Fig 9.The schematic of the RGB cube that illustrate both the primary and secondary colour at each vertex is shown below. Green and Blue) and the secondary colour (Cyan.6 40 . 9. there is difficulty in choosing the threshold as the histogram is no more bimodal This root cause is due to the variations of light or colours in the images. The cube vertex consists of the primary colour (Red. Magenta and Yellow). The colour space is normally graphically shown as RGB colour cube.3 Colour Image Processing RGB Colour Space The RGB image is made of colour pixels of an M x N x 3 array.

41 Fig 9.8 : Binary mage 41 .7 : Original Image Fig 9.

42 10 IMAGE ENHANCEMENT The aim of this process is to increase and improve the visibility of the image. black will becomes white and white will be come black . 42 . 10. contrast manipulation. For the complement of the true color image. binary or true color image. color image processing and image segmentation as well. A median filter is more effective than convolution when the goal is to simultaneously reduce noise and preserve edges. The pixel value obtained by the differences will be used in the output image. zeros will become ones or ones will become zeros. Image Enhancement techniques consists process of sharpening the edges image. the value of each pixel will be subtracted by the maximum pixel value which is supported by the class. the light areas will become darker and the darker area will become lighter.2 MEDFILT2 Median filtering is a nonlinear operation often used in image processing to reduce "salt and pepper" noise. 10. reducing noise. For the complement binary image.1 IMCOMPLEMENT Imcomplement will compute the complement image(IM). IM can intensity. Therefore.

43 11 Thresholding Image thresholding enjoys the central position of any application in the process of image segmentation. A more sophisticated approach might be to create a histogram of the image pixel intensities and use the valley point as the threshold. this may be computationally expensive. the mean or median will work well as the threshold. Representing F (x. The histogram approach assumes that there is some average values for both the background and object pixels. One of the simplest process will be the Gray-level thresholding. 11. However. often making the selection of an accurate threshold difficult. 43 . Several different methods for choosing a threshold exist. An initial threshold (T) is chosen. the rationale being that if the object pixels are brighter than the background. Fast and cheap. A simple method would be to choose the mean or median value. It is still greatly used in applying on the applications in the market due to certain reason. is the following iterative method: 1. they should also be brighter than the average. this will generally not be the case. and is robust against image noise. but that the actual pixel values have some variation around these average values. In such cases a unimodal threshold selection algorithm may be more appropriate.y) =1 for image object. F (x.y) = 0 for background of the object and T = threshold. Thresholding is one of the oldest segmentation method. One method that is relatively simple. or a thresholding algorithm can compute a value automatically. users can manually choose a threshold value. = 0 if K (x. however. and image histograms may not have clearly defined valley points. The regions of the images are classified by the reflectivity and absorption of light on its surface. It also transform the input image(K) into an output binary image (F) which is being segmented.y) = 1 if K (x. as mentioned earlier). The value of threshold (T) is being selected and compare with the pixel of the image.y)>=T.1 Threshold Selection The key parameter in the thresholding process is the choice of the threshold value (or values. does not require much specific knowledge of the image. In a noiseless image with uniform background and object values. which is known as automatic thresholding . this can be done randomly or according to any other method desired.y) < T. F (x.

The average of each set is computed. We can use the graythresh toolbox function to calculate the threshold value by using Otsu‘s method. 44 . f(m. The threshold defines the value of the gray level in which the baseline boundary is in between the foreground pixels and the background. keep repeating until the new threshold matches the one before it (i. Go back to step two. nth row) 3.3 Adaptive Thresholding The vehicle number plate can be partially illustrated.n) T} (background pixels) (note. The threshold value which is selected is use to reduce the intraclass variance between the white and black pixels. G2= {f(m. m1= average value of G1 2. If the histogram is bimodal.n):f(m. This iterative algorithm is a special one-dimensional case of the k-means clustering algorithm. For other cases. now using the new threshold computed in step four. m2= average value of G2 4. This is a common reason why global thresholding fail. in which the modes are overlapping.. the value of the threshold will be taken in the middle of the bimodal.n):f(m. G1= {f(m.e. its can solve the disadvantages of the global thresholding as it computes the value of the threshold for every pixel separately using the neighborhood.44 2. 11. By using the adaptive thresholding.2 Global Thresholding This is a method whereby the histogram of the image is being partitioned using a single threshold value.n) is the value of the pixel located in the mth column.n)>T} (object pixels) 2. The image is segmented into object and background pixels as described above. The value of the gray level varies from the darkest (0) to the brightest (255). T‘ = (m1+m2)/2 5. 11. until convergence has been reached). the threshold will try to reduce the error of taking the background pixels as object pixel and vice versa. 1. which has been proven to converge at a local minimum— meaning that a different initial threshold may give a different final result. creating two sets: 1. Image is form up of square matrixes called pixels and each pixels in the image have a brightness level known as gray level. A new threshold is created that is the average of m1and m2 1.

softening is often applied to the low pass filter and it will affect the image to become blurring. 12. It has been cropped to a rectangular shape in which it enhances the algorithm to be more uniformed in the further stages. We can apply convention techniques such as softening and sharpening.P). From the amplification of the signal. Thus.1 Filtering In this pre-processing stage. We will be enlarging the signal from the received light photons when we are using the digital camera with a higher ISO setting. filtering process is used for blurring and for noise reduction and thus resulting a more sharpen image. Noise in the images is appears as the colour speckles whereby there should not be any. In order to reduce the background noise. we have managed to obtain the black and white image of the Vehicle License Plate. blurring with liner/non liner filter is able to achieve noise reduction. 12. 12. Blurring is used in the preprocessing step which involved in removing and cleaning up of small patches and details from the image prior to object extraction. The syntax is as bellowed: BW2 = bwareaopen(BW. The reason of noise contamination on the image is categorised as below:  There will be not enough light source for proper exposure to activate in a low light area/ Therefore. we need longer time for the image sensor for collection of weak signal.45 12 IMAGE NOISE Digital camera with long exposure time setting and high ISOs will produced images with lots of noise. 45 . In this circumstance.3 FILTERING OF DIGITS Now. bridging of gaps in curves and lines. the electrical noise in the background is also being amplified. the electrical noise from the background will be higher than the signal. Next.2 Binary area open (Bwareaopen) The MATLAB toolbox function provide a bwareaopen function in which it removes the connected component which have fewer than P pixel. it will reproduce another binary image.

we need to identify the connected components. The pixel of 0 represents the background and pixel of 1 will make up the first object. Fig 12. we are able to remove or filter out the unwanted substances or noise that is not a character or digits. water droplet. Below illustrate an example of filtering process. the image is only left with characters and digits in which we are interested. After the component has been labelled.1 : Binary image with Noise For the above image. frame lines. so on and so for. By Filtering. 2) Identify the frame line that is connected to the digits and separate it. sand.46 The non-digits substance which appears in the image is caused by the noise (quantization that will transform to ‗Black‘). the pixel of 2 will label the second object. The elements of labels (stated in the algorithm) contains integer values that will be greater than or equal to 0. 12. 46 . object or substance which is not digits/characters such as screws. we use 8-connected (8 specify 8-connected objects). 141 connected components are found in the BW image. effects on JPEG compression. Lastly. The component can have a value of either 4-connected or 8-connected.4 Identify and remove away the small connected objects For this stage. In this algorithm. The 2 stages for the algorithm is as follows: 1) Remove out the small objects or connected components. they will obtain a unique number.

Therefore. Next. in which it will produce another binary image. components with less than 6000 is classify as non digits components and need to be taken away.47 Fig 12.2 : Binary Image after Filtering. we have apply the median filter and bwareaopen (Image Processing Toolbox) whereby it will remove all the connected components from the binary image that have value less than P pixels. 47 .

1 FEATURE EXTRACTION FROM DIGITAL IMAGE The digital image description is depends on the external and internal representation. vertical or diagonal edges etc. Secondly is the boundary-based segmentation. This will lead to a success in pattern recognition process. For this project. The normalized character description is based on the external characteristics as we only work for properties on the shape of the character. the descriptor of the character from the same class is close to each other in the vector space. Firstly is the global Knowledge of an image. Lastly will be the edge-based segmentation.48 13 Image Segmentation Image segmentation plays an important and critical step that lead to the analysis of the processed image data. n1=imagen(min(r):max(r). The process of the feature extraction is to transform the bitmap data into a form of descriptor in which more suitable for computer. end 48 . If we classify the similar character into classes. The process uses the edge detection to obtain the region contours and the objects will construct from the obtain contours. which is by using threshold. for n=1:Ne [r. The feature of the image is represented by a histogram. [L Ne]=bwlabel(imagen). In order to extract and analyzed the object characteristic. the process need to partition the image into different parts that will have a strong correlation with the objects. letter=read_letter(img_r). word=[word letter].min(c):max(c)).c] = find(L==n). The algorithm below will represent how the extraction is being carried out and extracted figure is illustrated as below. imshow(~n1). The descriptor vector includes the characteristics as the number of lines. we will be trying out first category segmentation method. 13. img_r=same_dim(~n1). Segmentation process can be categorized into several parts. disp(Ne). The color or texture of the image is basically the internal representation whereas the external representation is based on the characteristic of the shapes.

It will ignore the class if the image thresholding is too low. The database would be enlarged in order to improve the accuracy and better chances of obtaining the correct result. Database is needed to make sure that the image space can contained enough characters which have been extracted and the vehicle license plated number stored in the excel sheet for the purpose of comparison. If the class has match correctly. thus it can be accessed easily and updated. Database can be in the form of text. The class recognition will check if the computed ratio is correct.2 Database It is a collection of information or data which it is being orderly organize. 49 .49 Plotted Bounding Box Fig 13. contents and images.1 13. it will be compare with data set in the database and provide an output if the image is recognized correctly.

the extracted characters are resized to fit the characters into a window.2.50 14 Character Recognition 14. 14. 14. An algorithm for obtaining normalized cross correlation from transform domain convolution has been developed. several inexact but fast spatial domain matching methods have also been developed. The new algorithm in some cases provides an order of magnitude speedup over spatial domain computation of normalized cross correlation. In the expansion of d2 50 . Unfortunately the normalized form of correlation (correlation coefficient) preferred in template matching does not have a correspondingly simple and efficient frequency domain expression. Due to the computational cost of spatial domain convolution. each character is normalized to the size of (42x24) binary image and then follow by reshape to standard dimension before sending the data set to the next for comparison with the templates.1 Normalization In this phase. see Lewis .2 Template Matching The correlation between two signals (cross correlation) is a standard approach to feature detection as well as a component of more sophisticated techniques.v). Textbook presentations of correlation describe the convolution theorem and the attendant possibility of efficiently computing correlation in the frequency domain using the fast Fourier transform. For the project.y under the window containing the feature t positioned at u. For this reason normalized cross-correlation has been computed in the spatial domain.1 Cross Correlation The use of cross-correlation for template matching is motivated by the distance measure (squared Euclidean distance) . (where f is the image and the sum is over x.

51 .51 the term is constant. If the term constant then the remaining cross-correlation term is approximately is a measure of the similarity between the image and the feature.

the figure illustrate below lead to wrong recognition of the character. For instance. Fig 15.1 Extracting of Individual Digits Determine the angle of the Vehicle License Plate.52 15 ANALYSIS OF RESULT Tests are conducted according to the modular form 15.2 : Recognition: MH01AX7946 52 .1 Recognition : D6LH4CNEB800 Fig 15. It is generally important when capturing the vehicle image.

After much study the MATLAB toolbox syntax. This will result fail recognition of the vehicle license plate. Fig 15. Equalization and quantization allow to obtain a gray scale image with improve contrast between digit and the background. we can improve on the accuracy of capturing of vehicle image.53 To improve in the cropping of image.6000). Next. such as very dark image. figure. the binarized image do not allow to make the difference between the background and the digits. low contrast image. To improve on the performance of the character recognition. Below illustrate a example of the scenario: Morphologically open binary image (remove small objects) imagen = bwareaopen(imagen.2 Vehicle Quantization and Equalization For some rare cases. I have experience loosing out character when performing the extracting process.imshow(imagen). high contrast image.3 Checking and Verification of the Statistic In the midst of trying out other car plate. 15. 15. we have managed to make adjustment on the value and apply the skill of binary area open (BWAREAOPEN) in which help to remove the small object too.3 53 . we can make the difference between the digits and background inside the license plate. we can also include an algorithm where we can change the image of the four coordinates to a standard rectangle size or manually crop the image.

4 54 .imshow(imagen). figure.54 Morphologically open binary image (remove small objects) imagen = bwareaopen(imagen.7000). Fig 15.

Size . Using the mouse. 55 .Performs median filtering of the matrix A using the default 3by-3 neighborhood.Open file. 6.measures a set of properties for each labelled region. Rgb2gray . 9. 8.computes the complement of the image.Bwlabel . When you are finished sizing and positioning the crop rectangle.Regionprops . you specify the crop rectangle by clicking and dragging the mouse. The Crop Image tool is a moveable. zeros become ones and ones become zeros. 10. the pointer changes to cross hairs when you move it over the target image. 4.Morphologically open binary image (remove small objects). black and white are reversed.Global image threshold using Otsu's method.returns in num the number of connected objects found in BW. Imcomplement . based on threshold. 2. Imcrop . 5.Bwareaopen . Fopen . Graythresh . In the complement of a binary image. called the target image. create the cropped image by double-clicking the left mouse button or by choosing Crop Image from the context menu. or obtain information about open files. The output image BW replaces all pixels in the input image with luminance greater than level with the value 1 (white) and replaces all other pixels with the value 0 (black). Imread : Read image from graphics file. Medfilt2 .returns the sizes of each dimension of an image in matrix a and b.Convert RGB image or colormap to grayscale. 11. You can move or resize the crop rectangle using the mouse. imcrop returns the cropped image 7.55 16 List of functions used 1. When the Crop Image tool is active. 12. Im2bw .creates an interactive Crop Image tool associated with the image displayed in the current figure. 3. resizable rectangle that you can position interactively using the mouse.Convert image to binary image.

1] and Curvature [0.Smallest elements in array.2-D correlation coefficient.Disp .Determine whether array is empty.. 19. 20.Fclose . 16. without printing the array name.Display image.displays an array.0. 17.e.Largest elements in array. no curvature).Close one or more open files. 15.draws a rectangle with Position [0.Resize .Resize image.Find indices and values of nonzero elements. 18.Find .Isempty .56 13. 23.Min . 56 . 21.Corr2 .Fprintf . 14.1.0] (i.Rectangle .Imshow .Write text to device. 22.Max.

propied(n). image = medfilt2(image).txt'. title('INPUT IMAGE WITH NOISE') % Convert to gray scale if size(image. pause(1) image = imcrop(image). image=imread('exp4.'BoundingBox').'EdgeColor'.BoundingBox.1) rectangle('Position'.3)==3 % RGB image imagen=rgb2gray(image).title('binary image').imshow(~image). figure.threshold).jpg'). 'at'). hold on for n=1:size(propied. % Remove all object containing fewer than 6000 pixels image = bwareaopen(image.imshow(image). title('INPUT IMAGE WITHOUT NOISE') word=[].57 17 CODE 17.%Storage matrix word from image fid = fopen('log.%pause(10). end % Convert to binary image threshold = graythresh(image). [L num]=bwlabel(image).1 Algorithm to extract characters ( extractdigit. propied=regionprops(L.2) end hold off pause (1) 57 .'g'. % Show image figure.'LineWidth'.%Opens a text for append in order to store the number plates for log. % Show image binary image figure.6000).m). image =~im2bw(image.imshow(image). disp(num).

\nSee the log.min(c):max(c)).58 figure for n=1:num [r.3 Algorithm to read letters (read_letter.'Number Plate:-%s\nDate:-%s\n'.[42 24]). letter=read_letter(img_r). 17.\nExtracted Number plate:. elseif vd==3 58 . 17.word). end vd=find(comp==max(comp)). msgbox(sprintf('Number Plate Extraction successful. word=[]. figure.5) end fprintf(fid. pause(0.2 Algorithm of Normalisation of characters (same_dim.%s . if vd==1 letter='A'.imagn).txt file to see the stored number.n}. n1=image(min(r):max(r).m) function img_r=same_dim(imagen_g) img_r=imresize(imagen_g.'Extraction Success').imshow(~n1). load templates for n=1:36 sem=corr2(templates{1.word. fprintf(fid. word=[word letter].date). elseif vd==2 letter='B'.'------------------------------------\n').'. img_r=same_dim(~n1). fclose(fid). comp=[comp sem].m) function letter=read_letter(imagn) comp=[].c] = find(L==n).

elseif vd==4 letter='D'. 59 . elseif vd==18 letter='R'.59 letter='C'. elseif vd==19 letter='S'. elseif vd==9 letter='I'. elseif vd==11 letter='K'. elseif vd==12 letter='L'. elseif vd==20 letter='T'. elseif vd==14 letter='N'. elseif vd==7 letter='G'. elseif vd==13 letter='M'. elseif vd==10 letter='J'. elseif vd==23 letter='W'. elseif vd==6 letter='F'. elseif vd==8 letter='H'. elseif vd==17 letter='Q'. elseif vd==15 letter='O'. elseif vd==5 letter='E'. elseif vd==21 letter='U'. elseif vd==24 letter='X'. elseif vd==16 letter='P'. elseif vd==22 letter='V'.

%*-*-*-*-* elseif vd==27 letter='1'. else letter='0'. elseif vd==29 letter='3'. elseif vd==26 letter='Z'. elseif vd==31 letter='5'. elseif vd==34 letter='8'. elseif vd==30 letter='4'. elseif vd==33 letter='7'. end 60 . elseif vd==35 letter='9'. elseif vd==28 letter='2'. elseif vd==32 letter='6'.60 elseif vd==25 letter='Y'.

61 18 Result Analysis and Discussion 18.1 Experimental results Extracted Number plate: MH12EM2587 Extracted Number plate: MH12FF7461 Extracted Number plate: MH12DM3909 61 .

62 Extracted Number plate:LMH12CD9736 Extracted Number plate:LK00A53N6523 Extracted Number plate: MH12HN0389 62 .

Perhaps the most important conclusion reached from this study is that the utility of ALPR depends on volume. The initial analysis of parking lot data produced by Schuurman (2007) suggested that parking lot 63 . The technology may have the capacity to read more plates.000 plates per hour.2 Discussion Although ALPR technology has been promoted as being capable of reading in excess of 3.63 Extracted Number plate: MH14CS7710 Extracted Number plate:MH12DE94 Extracted Number plate: MH01AX7946 18. but traffic volume and the design of Surrey streets may impede this read rate. the current results suggest that this capability is significantly overstated in actual road tests.

the current study results also indicated that stolen vehicles represented a very small proportion (approximately one per cent) of all licence plate hits. Given this. for the most part. for a stolen vehicle to be detected by an ALPR-enabled vehicle. at least. in order to maximize efficiency. However. Second. as suggested above. Despite the promise that ALPR technology holds for catching stolen vehicles. First. along one of the four traffic corridors assigned for that day. the greater the number of raw hits. An important consideration raised by this research has to do with prioritizing responses and the allocation of scarce resources. it may be more beneficial to use stationary cameras along a large number of intersections. considering the results of this phase of the project. for the most part. if the objective is to identify stolen vehicles as this will. In other words. the fact that officers could expect several hits per hour requires the design of a response priority scheme and increase patrol units to manage the increased workload. the more cars scanned. it may be possible. In effect. 64 . As a consequence. Importantly. the nature of hits was basically uniform for all of the assigned traffic corridors and the proportions held when considered by time of day or day of month. However. and. and at the exact location where a moving ALPR-enabled vehicle could photograph its licence plate. the frequency of hits exceeded what a typical patrol unit could respond to during a shift. If patrol officers are faced with a significant increase in the number of hits as a result ALPR deployment. to train volunteers who can assist the police in filtering through database hits. There are two main reasons why ALPR. A large advertising campaign can inform the public about the nature and use of this technology which may result in the public being less willing to violate traffic regulations. the hit list that the ALPR-enabled vehicle searched against was 24 hours old. until such time as ALPR-enabled vehicles have access to real-time data and there are many more ALPR-enabled police cruisers on the road at the same time. Rather. it was all about the number of hits. the stolen car had to be on the road in Surrey.64 deployment was dependent upon the number of vehicles in parking lots and. as currently operated in Surrey. at least 24 hours after the car had been stolen. Unfortunately. remove the variable of a roving ALPR which likely reduces success. police forces operating ALPR technology may want to focus on high volume traffic corridors during the day shift. was not effective in locating stolen vehicles. the same conclusions applied to this road test of the technology. this result should not have been unexpected. substantiating which calls are valid and which must receive priority attention to lessen the burden on patrol officers. these statistics were not improved in the current analysis. In both cases. detachments must design response schemes to prioritize which hits officers respond to. ALPR requires a meeting in time and space between the police cruiser and the stolen vehicle. The likelihood of this scenario occurring is extremely low.

usually because the plate is too far away but sometimes resulting from the use of a low-quality camera. particularly motion blur. However. or dirt on the plate. When running the main. Australia. tow bars may obscure one or two characters of the license plate. though in some countries and jurisdictions. Bikes on bike racks can also obscure the number plate.m file. the decision may be made to have an acceptable error rate of one character.65 19 Difficulties and Recommendations There are a number of possible difficulties that the software must be able to cope with. Blurry images. Increasing the height of the camera may avoid problems with objects (such as other vehicles) obscuring the plate but introduces and increases other problems. An object obscuring (part of) the plate. such as Victoria. Lack of coordination between countries or states. Some small-scale systems allow for some errors in the license plate. Poor lighting and low contrast due to overexposure. popular for vanity plates (some countries do not allow such plates. For the future works and suggestion on improvements. 65 . this level of inaccuracy would not be acceptable in most applications of an ANPR system. On some cars. Circumvention techniques. A different font. reflection or shadows. such as the adjusting for the increased skew of the plate. eliminating the problem). Two cars from different countries or states can have the same number but different design of the plate. quite often a tow bar. "bike plates" are supposed to be fitted. these are the steps which are recommended:  Modification is needed to be done on the offset of detecting the rectangular plate or by applying other technique to the system. When used for giving specific vehicles access to a barricaded area. These include:        Poor image resolution. it is able to detect and recognize some of the zoom in car vehicle license plate but fail on the image of cars with headlights. This is because the likelihood of an unauthorized car having such a similar license plate is seen as quite small. it is primarily left to the hardware side of the system to work out solutions to these difficulties. While some of these problems can be corrected within the software.

we can use RGB to HSV or RGB to CMY method to deal with other type of colour license plate.  There should be improvement on the decision of the algorithm and ways to detect error. When the probability of recognition guess is correct but falls below the threshold. 66 . the recognition system should refuse to make the decision. For future implementation.66  We only use the black and white image of the Vehicle license plate.

recognition of vehicle license plate characters. method of segmentation and recognition of vehicle license plate character and writing of simulation program. After doing much reading and research. research and with the help and understanding from the tutor. During the initial project and development stage. 67 . Next. After doing much research and absorbing more knowledge into MATLAB toolbox. problems were resolved by using the suitable syntax and methods. The problem encountered during the progress of the project include the selection of suitable method such as : sourcing of car images. the problem encountered is the selection of software program. we chose MATLAB software as it is easier to implement. methods on detection.67 20 Critical Review and Reflection The development of the project is considered quite successful although we still faced some problems during the progress of the project we managed to solve the problem by doing more reading. segmentation. we also encountered problems in detection.

Given that officers cannot check all the licence plates they encounter while on shift. Research in the United Kingdom suggested that in responding to the sheer number of hits identified through ALPR. untoward. detachment-specific schemes may need to be developed. While officers currently employ a set of indicators. it is highly plausible that through this process. This increases the potential for the recovery of stolen goods as well as convictions. they are compelled to make a series of decisions concerning which plates to search. in fact. Finally it is proved to be %96 for the segmentation of the characters and %80 for the recognition unit accurate. however. identified through experience. namely. giving the overall system performance 92. This system can be redesigned for multinational car license plates in future studies. The ALPR technology offers several benefits to police forces. police are better able to identify more persons of interest.57% recognition rates. as the profile of crime is different in jurisdictions. Firstly we extracted the plate location. With an increased number of ―hits‖. This system is designed for the identification INDIAN license plates and the system is tested over a large number of images. ALPR also offers an objectivity that may benefit police. As a consequence. but to other calls for service. The use of ALPR technology might also result in safer police driving as officers would no longer have the added distraction of turning away from the road periodically to manually type in licence plates of interest . with which to select those plates that appear more suspicious. some limitations inherent in the use of such advanced technology. such as having an officer physically type in a licence plate to scan. There are.68 21 Conclusion & Future Scope In this report. Research with ALPR has shown several benefits. increased police efficiency. then we separated the plate characters individually by segmentation and finally applied template matching with the use of correlation for recognition of plate characters. police must develop strategies that enable officers to prioritize their responses. impeding their ability to efficiently respond not only to ALPR hits. ALPR has the ability to quickly and efficiently scan a large number of licence plates without any officer intervention. The ability of ALPR to scan a large number of plates allows for more plates to be scanned faster and more efficiently. and unlicensed drivers much more quickly than previous police strategies. Most importantly. With the potential for an increased number of hits. In other 68 . However. or successful matching between a scanned plate image and a database of interest. officers could become overwhelmed by the increase in the number of problematic cars to respond to. prohibited drivers. officers may miss plates that are. The technology also allows the police to identify uninsured vehicles. an officer‘s workload substantially increased. we presented application software designed for the recognition of car license plate.

They noted that one company sells a clear spray (US $30 per can) that the manufacturers claim can make licence plates invisible when read by a camera. Deisman (2003) identified that there are limits to the extent that police in Canada can engage in continuous and non-selective monitoring of citizens. other criminal justice agencies. privacy concerns are also a limitation to the use of ALPR technology. more recently. without the successful coordination between police forces and agencies. For instance. recognition software has produced extremely low successful recognition rates (Gordon and Wolf. or drivers who are uninsured. to better respond to priority hits. Citizens may equate the use of ALPR technology to ―fishing expeditions‖. those using ALPR technology must have access to data that details information about stolen plates or cars. in the event that a plate is not read correctly and the officer deems the plate suspicious. as opposed to specifically searching out particular plates based on prior intelligence. The efficiency of ALPR technology is entirely dependent upon the successful coordination of agencies. there will be those who develop the means to defeat these techniques. police forces may find it necessary to increase the number of officers on the road which. given current fiscal realities in many detachments in Canada. Gordon and Wolf (2007) reported that since the advent of ALPR. ALPR technology simply will not succeed. Concerned citizens may accuse police or the government of using the technology to track law-abiding citizens. Information can either be provided in real time. it also provides new methods with which to avoid being screened by such technology. Much of the work involved in implementing ALPR technology involves building these initial relationships between agencies.69 words. Lastly. lists are updated as cars or plates are reported stolen or it can be updated every 24 hours. In the past.g. 2007). and the motor vehicle branch. However. ALPR cannot possibly identify plates of interest. such as licence plate recognition. Moreover. 2007). it is important that officers to retain the ability to investigate the plate more fully and to follow their instincts when observing a suspicious vehicle. depending on the geographic location of hotspots. Yet. where police simply scan all plates until they get a hit. e. Schuurman 69 . Concerns in Canada already exist regarding the use of Closed Circuit Television (CCTV) systems in public (Schuurman. While advances in technology allow for the successful use of systems. Without the provision of data by which to compare scanned plates. drivers that have been prohibited or have lost their licence. and the specific needs of the community. Again. Either way. may not be feasible. research suggested that the ALPR technology reads plates correctly 95% of the time. the number of officers on patrol. as the police develop new technologies. some companies have begun to sell products to thwart the technology. priority schemes may need to be individualized. such as insurance companies. invading their right to privacy. vehicles that have been involved in other criminal activity. In effect.

Future Scope Though we have achieved an accuracy of 80% by optimizing various parameters. there may be concerns regarding who has access to this data .70 (2007) also noted that the Canadian Charter of Rights and Freedoms states that privacy rights of citizens are breached by indiscriminate video surveillance without cause. The overriding benefit of ALPR is that it brings a far larger number of offenders to the attention of the police. ALPR does have a specific utility as its strategic deployment will assist police departments to more effectively response to a variety of auto and driving-related offences. 70 . weekly. How to adequately respond to this situation will require careful thinking and planning on the part of the police. Still. although further research is needed to determine the extent to which ALPR increases the rate of arrest and has a deterrent effect. it is required that for the task as sensitive as tracking stolen vehicles and monitoring vehicles for homeland security an accuracy of 100% cannot be compromised with. the results of this study suggest that ALPR technology offers several substantial benefits to the police. and thus make any firm conclusions on its general utility. This work can be further extended to minimize the errors due to them. Therefore to achieve this. or yearly basis. further optimization is required. In addition. monthly. it is difficult to assess the impact of ALPR on police resources and workloads. Citizens may also have concerns with respect to the maintenance of data in warehouses. It is. extremely important that considerable thought is given to the safe storage of data and strict regulations regarding who has access to the databases. Also. In conclusion. more research needs to be conducted in order to determine how ALPR technology can be balanced with respect to citizens right for privacy and civil rights. Essentially. therefore. the issues like stains. In responding to concerns of privacy. . policies may be put in place that regulate the deleting of collected data on a daily. blurred regions & different font style and sizes are need to be taken care of. smudges. Citizens may fear the potential for breaches in security. rather than the few offenders the police are able to find during their routine activities. However.

71 Annexure A: Project Plan & Progress Report For developing the project. Task 3 Construct Matlab software for simulation (24 Feb 2012 to 20 Mar 2012)  Writing and create algorithm for extraction of vehicle license plate character  Writing and simulate the block by integrating all the written algorithm together  Version R2008a for editing. we have drafted out the project schedule to follow up closely. Thus. Task1 Research (24 Jan 2012 to 10 Feb 2012)        Research on various methods used for Vehicle License Plate Recognition Conversion of Images Image Thresholding Noise filtering Vehicle License Plate Detection Vehicle License Plate Extraction/Segmentation Vehicle License Plate Recognition Task 2 Development (11 Feb 2012 to 23 Feb 2012)  Determine the value of pixel need for filtering  Determine algorithm for plate detection  Determine algorithm for extraction/segmentation  Determine algorithm for plate recognition. There is interconnected link between the research work and the designing and Implementation process.  Fine tuning of the integrated block algorithm Task 5 Reporting Writing (28 April 2008 to 1 May 2012)  Allocate at least 1 week for writing the final report. Task 4 Testing and Fine Tuning (30 March 2012 to 24th April 2012)  Fine tune on the filtering and calculation of connected components. it is important to choose the type of model and syntax that are suitable for Matlab is taken into consideration. 71 .

Irwin M. AND BREANNE CAVE .ht ml  Matlab software version R2008a. It is a very user friendly device as it will guide and explain to us on the specific toolbox in which we will be using. Author ONDREJ MRTINSKY  Lecture notes  Mathworks. ― International Edition Digital Image Processing Second Edition‖  John C. Woods and Steven L. Cohen.72 References  Rafael C. McCormick SCHOOL of CRIMINOLOGY and CRIMINAL JUSTICE University College of the Fraser Valley.Hopgood.Edward B.com/access/helpdesk/help/toolbox/nnet/nnet_product_page.‖Intelligent Systems for Engineers and Scientists‖  William J.ht ml#printable_pdf.com/access/helpdest/help/techdoc/mathlab_product_page2. ―Digital ImageProcessing using MATLAB‖  Rafael C. Gonzalez and Richard E. and Amanda V. Center for Evidence-Based Crime Policy. JULIE WILLIS. LICENSE PLATE RECOGNITION TECHNOLOGY (LPR) IMPACT EVALUATION AND COMMUNITY ASSESSMENT .  Dr. Richard E. Russ. Gonzalez. Darryl Plecas. George Mason University . ―The Image Processing Handbook‖  Adrian A.  Mathworks. LINDA MEROLA. Eddins. A Report on the Utility of the Automated Licence Plate Recognition System in British Columbia 72 .  CYNTHIA LUM.‖An Engineer‘s Guide To Matlab‖  ALGORITHMIC AND MATHEMATICAL PRINCIPLES OF AUTOMATIC NUMBER PLATE RECOGNITION SYSTEMS. Dr. ―Introduction To Matlab 7 For Engineers  Magrab. Woods. Palm III.

3 1 Aug-2 Sept.platerecognition. J.‖ Proceedings of the IEE Colloquium on Image analysis for Transport Applications.G.‖ IEE Colloquium on Electronic Image and Image Processing in Security and Forensic. Irecki. Bailey. Fahmy. L. Bird. 1990.A. Venema.org/wiki/Vehicle_registration_plates_of_India  http://en.Serkan Ozbay. V01. K.anpr. Proceedings of First IEEE International Workshop on Electronic design.P. Helmholt. M.035. Ter Brugge. Yang ―Test bed for number plate recognition applications‖.A. 1995. 2002.S. Aprl.A. Pluim. A. pp.  http://en. Lim and L.wikipedia. ―Number-plateMatching for Automatic Vehicle Identification. Johnson.S.wikipedia.info/ 73 .org/wiki/Automatic_number_plate_recognition  http://www. Westenberg.H.  R.D. Spaanenburg. B.W.G.K. Test and Applications ( DELTA‘02 ). Nijhuis. ―Car License Plate Automatic Vehicle Identification by Plate Recognition Author .M. 1990.73  D. B. 1994  J.M. and Ergun Ercelebi. D. 1990.  A. ―Automatic Number-plate Recognition : Neural Network Approach.M. and AS. IEEE Computer Society. February 16. Lotufo. M.net/  http://www.  M. 1990.6/1-6/6.‖ Proceedings of VNIS‘94 Vehicle Navigation and Information System Conference. Morgan.A. R. ―Automatic NumberPlate Recognition. Johnson. 1994.

74 74 .