Vehicle License Plate Location Based on Improved Roberts Operator and Mathematical Morphology

Abstract

The License Plate Location(LPL) from the vehicle images is the key step in the Automatic License Plate Recongnition (ALPR). This paper proposes a new algorithm for vehicle license plate location based on improved Roberts operator combining with mathematical morphology. Traditional Roberts operator is improved by added two templates in other directions. Then image edges are detected by selecting the appropriate threshold for Roberts operator and mathematical morphological analysis is utilized to get connected regions which contain the candidate plate region with appropriate size. Finally analyze candidate regions by connected component and position plate region. Experiments have been implemented on images taken from different background: lightening conditions (night and day), license angles, illumination, reflected light, dynamic conditions, The experimental results show that this method is effective in quickly and reliably detecting and locating the vehicle license plate region.