21-05-2013, 03:38 PM
Vehicle License Plate Detection Method Based on Sliding Concentric Windows and Histogram
Vehicle License Plate.pdf (Size: 1.07 MB / Downloads: 57)
Abstract
Detecting the region of a license plate is the key
component of the vehicle license plate recognition (VLPR)
system. A new method is adopted in this paper to analyze
road images which often contain vehicles and extract LP
from natural properties by finding vertical and horizontal
edges from vehicle region. The proposed vehicle license plate
detection (VLPD) method consists of three main stages:
(1) a novel adaptive image segmentation technique named
as sliding concentric windows (SCWs) used for detecting
candidate region; (2) color verification for candidate region
by using HSI color model on the basis of using hue and
intensity in HSI color model verifying green and yellow
LP and white LP, respectively; and (3) finally, decomposing
candidate region which contains predetermined LP
alphanumeric character by using position histogram to
verify and detect vehicle license plate (VLP) region. In
the proposed method, input vehicle images are commuted
into grey images. Then the candidate regions are found by
sliding concentric windows.
INTRODUCTION
With the rapid development of highway and the wide
use of vehicle, people start to pay more and more attention
on the advanced, efficient and accurate intelligent transportation
systems (ITSs). The task of recognizing specific
object in an image is one of the most difficult topics in
the field of computer vision or digital image processing.
VLPD is also very interesting in finding license plate area
from vehicle image. The vehicle license plate detection is
widely used for detecting speeding cars, security control
in restricted areas, unattended parking zone, traffic law
enforcement and electronic toll collection. Last few years
have seen a continued increase in the need for and use
of VLPR. The license plate detection is an important
research topic of VLPR system. Because of different
conditions such as poor illumination and varied weather, it
is important and interesting how to segment license plate
fast and perfectly from road images which often contain
vehicles.
REVIEW OF OTHER METHODS
This section provides a descriptive summary of some
methods that have been implemented and tested for
VLPD. As far as detection of the plate region is concerned,
researchers have found many methods of locating
license plate. For example, a method based on image segmentation
technique named as sliding windows (SW) was
also proposed for detecting candidate region (LP region)
[1], main thought of image segmentation technique in
LP can be viewed as irregularities in the texture of the
image and abrupt changes in the local characteristics of
the image, manifesting probably the presence of a LP.
Color arrangement of the plate
Assorted styles of license plates found on vehicles in
South Korea are shown in Table I. Each style is associated
with a particular class of vehicle. The classes include private
automobile, taxi, truck, bus and government vehicles.
Each style has a different foreground (character) and/or
background (plate) color. However, in all only five distinct
colors (white, black, green, yellow and deep blue) are
utilized in these license plates. We shall pay attention to
three different plate colors when searching for LP in an
input image. Color arrangements for the Korean VLPs are
shown in Table I.
Outline of the Korean VLP
Standard LP contains Korean alphabets and numbers
which are shown in Fig. 1. Few LP contains Korean
alphabets and numbers in two rows, in future this kind of
LP is supposed to be converted into a single row. Where
plate color is white and character color is black, they
contain seven alphanumeric characters written in a single
line. Fig. 1 shows, where plate color is green and character
color is white, they contain Korean LP in two rows. The
upper row consists of two small Korean characters of
region name followed by one or two numbers of class
code or two numbers of class code and a usage code
of one syllable (Korean character). The lower row is
usage code of one syllable (Korean character) and four
big numbers serial code or only four big numbers to
indicate the usage and serial number respectively. When
plate color is yellow and character color is black, some LP
contains all alphanumeric characters written in a single
line, and another type of yellow LP is found and they
contain Korean LP in two rows.
PROPOSED ALGORITHM
In the author’s previous work [5], a parallelogram and
histogram based vehicle license plate detection (VLPD)
was presented. We propose in this section an enhanced
version of VLPD algorithm. As shown in Fig. 4, the
extracted features using a recursive algorithm is implemented
for connected component labeling (CCL) operation
and during this step main geometrical property of LP
candidate such as aspect ratio is computed. This parameter
is used to eliminate LP-like object from candidate
list. In addition, we are segmenting each predetermined
alphanumeric character after using position histogram.
Authenticating candidate region color
Many applications use the HSI color model. Machine
vision uses HSI color space in identifying the color of
different objects. Image processing applications such as
histogram operations and intensity transformations are
performed on an image in the HSI color space.
The RGB color space consists of the three additive
primaries: red, green and blue. Spectral components of
these colors combine additively to produce a resultant
color. RGB model is represented by a 3-dimensional cube
with red, green and blue at the corners on each axis.
Character extraction
Information extracted from image, intensity histograms
play a basic role in image processing, in areas such
as enhancement, segmentation and description. In this
section, verification and detection of the VLP region
as well as character segmentation, are considered and
discussed in this study.
Once the candidate area is binarized the next step is to
extract the information. At first, regions without interest
such as border or some small noisy regions are eliminated,
the checking is made by height comparison with other
plate characters height.
CONCLUSION
In this paper, we adopt a new method in image segmentation
technique named as sliding concentric windows
(SCWs) based on extract candidate regions by finding
vertical and horizontal edges from vehicle region; and
then color verification for candidate regions by using
HSI color model on the basis of using hue and intensity
in HSI color model verifying green and yellow LP and
white LP, respectively. Finally, candidate region which
contains predetermined LP alphanumeric character by
using position in the histogram to verify and detect vehicle
license plate (VLP) region was decomposed.