Research on Image Recognition for License Plate from Classifier Algorithm

Abstract

Classification algorithm based on Adaboost and Harr-like features is quite mature to be used in the field of face detection. In this paper, the traditional Adaboost algorithm is improved and the vehicle positioning module is added. At the same time, this algorithm is applied to vehicle license plate location and it greatly improves the accuracy of license plate location. License plate location is to distinguish the plate under the complex background environment. In license plate recognition, it is the most critical step. Its accuracy directly affects the accuracy of license plate recognition. In this aritical, a new algorithm is proprosed, and two simulation softwares are combined to make results of license plate location and recognition accuracy improved significantly.