License plate localization and recognition under different illumination conditions

Abstract

License Plate Recognition (LPR) is a key technology for many applications. But, most methods in the literature for LPR are only suitable for stable illumination environment. In this paper, we present an algorithm for both license plate localization and recognition in day time and night time. The first step of the proposed method is to enhance the image using high dynamic range (HDR). The tile correction and an adaptive segmentation based on image quality are applied to obtain upright characters. Finally, the combination of three well-known classifiers is used to achieve character recognition.