Automated vehicle license plate detection system using FRIT Algorithm

Abstract

expensive and slow. In this thesis a novel technique has been developed for the license plate detection system using Finite Ridgelet Transform (FRIT) algorithm. The proposed technique is simple and effective which outperforms other existing methods for License Plate Detection (LPD). To verify the effectiveness of the proposed technique, experiments have been conducted using benchmark vehicle images. This gives remarkable success of detection in real time processing. Experiments result shows that the proposed technique is capable of detecting license plate successfully in various lighting condition. This technique is completely language, font, color and size independent. It is rotation invariant and capable of detecting license irrespective of poor illumination condition, multiple license plates in a same image and low quality noisy image. It is also effective in the shadow condition or the plate image is blurred. Achieved from the proposed technique the results have been compared with those off other research and shows that our technique. We also compare the results and efficiency of this work with other existing systems/techniques which gives better result.