Support Vector Machine Method to Reduce the Execution Time of Vehicle Plate Recognition System

Abstract

This research aims to create a vehicle plate detection and recognition system with Cascade Classifier, Support Vector Machine (SVM) and Optical Character Recognition (OCR). Cascade Classifier with Local Binary Patterns (LBP) descriptor is used to detect the car licence plate (Coarse Location). SVM is used to reduce plate candidate detection error and the execution time. Optical Character Recognition (OCR) is used to recognize the characters in plates. The system test is performed using 19 video data of moving vehicles at night and rain conditions. Each video has a duration of 30 seconds and contains 4-10 cars per video. The testing results show reduction of the execution time of vehicle plate recognition system which reached 60% with the average accuracy of plate recognition 61.94%.

Comparative Study between Iraqi and EU Vehicles License Plates in view of Segmentation and Recognition

“…The features obtained from undecimated coefficients are used to recognize plates. Ismail et al in [8] used Cascade Classifier, SVM and OCR in the process of detection and recognition. The aim of its proposed system was to reduce detecting and recognition execution time while vehicle moving at noisy conditions.…”