A Technical Assessment on License Plate Detection System

A Technical Assessment on License Plate Detection System

Jeena Rita K. S. (SCMS School of Engineering and Technology, India) and Bini Omman (SCMS School of Engineering and Technology, India)
DOI: 10.4018/978-1-5225-0889-2.ch009
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Identifying the region of interest from an image is an important task in the field of computer vision. This process is referred as Object Detection. Locating the object can be done by extracting the features from the image and the features depend on the application. Image retrieval and surveillance are two important applications of Object Detection. Surveillance is an active research topic in computer vision that tries to detect, recognize and track objects over images. An interesting application of it is License Plate Recognition module in Intelligent Transportation System. The speedy developments in economic and social life bring on for a large increase in the number of vehicles in the city. This makes the traffic management difficult. It has great impact on human life as it aims to increase the transportation safety through innovative technologies. An important function that need for majority of ITS application is to identify the vehicle using LPR. This chapter presents assessment on different methods in detecting the license plate and discusses a case study on it.
Chapter Preview
Top

Introduction

Object detection is the process of discovering the instances of real-world objects in images or videos. Object detection algorithms use mined features and learning algorithms to recognize instances of an object (Gossian & Gill, 2014). Human can easily recognize the objects in a scene but not for the computer. It is difficult to implement the algorithms that identify the object. Many difficulties are there for detecting the object from an image. View point variation, illumination, occlusion, scale change, deformation or articulation, background clutter are some of the challenges that face in object detection task. In general approach, object detection methods first learn from positive and negative samples that which all images have the particular object. Positive samples are the samples that contain the object of interest and negative samples are that do not possess object of interest. After learning from these samples it determines the object of interest from the new image. Object detection has application in the areas like image retrieval, surveillance, vehicle navigation and the like. Surveillance is an active research topic in computer vision that tries to detect, recognize and track objects over images. An interesting application of it is License Plate Recognition (LPR) module in Intelligent Transportation System (ITS). Here the object of interest is License Plate (LP).

The quick economic development and social progress pay way for a large increase in the number of vehicles in the city. This makes the traffic management difficult (Dong & Feng, 2014). ITS plays an important role in the field of traffic security. It has great impact on human life as it aims to increase the transportation safety through innovative technologies. Arterial management, Emergency management, Traveller information, Electronic payment, Road weather management, Driver assistance system, Collision notification system are some of the key applications of ITS. An important function that need for majority of ITS application is to identify the vehicle. For this LPR system is requiring. As the efficiency of LPR increases the ITS applications outputs good result. So it becomes a vital slice in Intelligent Transportation System.

Detecting the number plate region and recognizing the characters of the number plate is an interesting. It is technique that identifies the vehicles by their license plates. The system uses a camera to capture the front or rear of the vehicle which contain the LP. Then the LPR system processes that image and extracts the information. This information is then used for different applications. The main applications of LPDS come in: parking areas, to detect the vehicles that are in over speed, control in restricted areas, electronic toll collection etc. Licence plate is like a unique identifier (Comelli et al., 1995). No two vehicles have the same license plate. So there is no need for additional hardware fitted to the vehicle to hold its identity. The difficulties in detecting the license plate detection systems faces are poor lighting due to shadows, weather conditions, the image quality, plate model and color, processing time etc.

With the popularity of Intelligent Transportation System, automatic identification of vehicle becomes an important area. Generally, two ways are there to identify a vehicle. One is through electronic devices and other through image processing techniques. In the case of electronic devices, RFID tag is attached to the vehicle and using this identification is performed. An external antenna sends out microwave signal and this signal extract the unique information from the tag. With this information the vehicle identification process is performing. In the other case, camera catches the image of a vehicle and the LPR module extract the number plate information which is unique for each vehicle.

Complete Chapter List

Search this Book:
Reset