Automatic License Plate Recognition (LPR) System Using Cascade Forward Neural Network

Abstract

Automatic Vehicle Identification (AVI) has several applications in traffic systems (highway
electronic toll collection, red lightweight violation social control, border and customs
checkpoints, etc.). LPR is a good kind of AVI systems. Image pre-processing is completed to
form the input image suited additional process. During this project, a wise and easy formula
is planned for vehicle’s vehicle plate recognition system. The planned formula is three fold:
Extraction of plate region, segmentation of characters and recognition of plate characters.
For extracting the plate region Connected element Extraction (CCE) algorithms is employed.
For segmentation part Maximally Stable Extremal Regions (MSER) algorithms is used. And
finally Cascade Forward Neural Network (CFNN) is used to recognize the characters by
using Edge Orientation Histogram (EOH) and Gray Level Co-occurrence Matrix (GLCM)
features. The performance of the proposed algorithm is tested with set of 24 images. Based on
the experimental results, it is observed that the proposed algorithm outperforms the other
existing techniques.