Information Technology Reference
In-Depth Information
ITS deploys SBSs to build a fundamental transportation monitoring network for
real-time and quick-response trac applications. A SBS could capture various
kinds of vehicle information including license plate information. The challenge is
that Automatic License Plate Recognition (ALPR) cannot satisfy accuracy re-
quirements of ITS due to changeable weather and machinery malfunction among
the large-scale deployment SBSs [9,10]. Moreover, it is hard for ALPR to get a
quick response and high performance under the circumstances of real-time and
large-scale trac datasets.
In view of these challenges, a heuristic license plate correction method, named
HelpMe, is presented in this paper to improve the accuracy of ALPR. Unlike
ALPR, creative heuristic license plate inspection and correction algorithms are
applied in HelpMe and are deployed on HANA cluster. Therefore, HelpMe ob-
tains a real-time performance and high accuracy compared to ALPR.
The remainder of the paper is organized as follows: preliminary knowledge
about our method is presented in Section 2. Then a heuristic license plate cor-
rection method, named HelpMe, is described in Section 3 in detail. In Section 4,
experiments are designed and analyzed to evaluate the veracity and time con-
suming of HelpMe. Related works are presented in Section 5. Finally, the paper
is concluded and an outlook on our future work is discussed in Section 6.
2 Preliminary Knowledge
In-Memory Database.
An In-Memory database (IMDB) is a database manage-
ment system that primarily relies on main physical memory for computer data
storage compared with conventional database management systems (Disk Res-
ident Database, DRDB) that employ a disk storage mechanism [11]. The key
difference is that in IMDB the primary copy lives permanently in memory. As
a consequence, IMDB can provide much better response times and transaction
throughputs compared to DRDB [12]. In this paper, HelpMe is proposed to im-
plement on high performance analytics appliance database platform (HANA) to
meet the acquirement of real-time license plate correction [13]. The SAP HANA
platform is a flexible data source agnostic in-memory data platform that allows
users to analyze large volumes of data in real-time.
Automatic License Plate Recognition.
Automatic license plate recognition
(ALPR) is a mass surveillance method that uses optical character recognition
on images to read vehicle registration plates [14,15]. Typically, ALPR is used in
road-rule enforcement cameras or closed-circuit television, as a method of elec-
tronic toll collection on pay-per-use roads and cataloging the movements of trac
and individuals for various police forces. ALPR commonly uses infrared lighting
to allow the camera to take the picture at any time of the day. Many primary
algorithms implemented to identify a license plate from the images captured by
the cameras. However, no one algorithm can guarantee the absolute accuracy.
Existing systems have so many problems although lots of improvements in image
recognition have been made to decrease error rate [16,17].
