License Plate Detection Algorithm Based on Gentle AdaBoost Algorithm with a Cascade Structure

摘要

This paper aims at automatically detection of car license plates via image processing techniques. The method used is a so-called gentle AdaBoost algorithm which is combined with a cascade structure. The gentle AdaBoost (GAB) algorithm is known to have a higher detection rate and a lower false positive rate than the basic discrete AdaBoost (DAB) which is currently reported being used for the license plate detection. The use of a cascade structure in the gentle AdaBoost algorithm saves the computation time. Several testing results are provided and the comparisons with other algorithms are made.