Vehicle License Plate Tilt Correction Based on the Straight Line Fitting Method

Tilt correction is an integral part of the automatic vehicle license plate recognition (VLPR) system. In this paper, according to the least square fitting with perpendicular offsets (LSFPO), the VLP candidate region is fitted to a straight line. After the line slope is obtained, rotation angle is estimated. Then the whole image is rotated for tilt correction by this angle. Despite the success of VLP detection approaches in the past decades, a few of them can effectively locate license plate (LP), even when vehicle bodies and LPs have similar color. A common drawback of color-based VLP detection is the failure to detect the boundaries or border of LPs. To overcome a common drawback, in this paper, the authors propose a modified recursive labeling algorithm for solving this problem and detecting candidate regions. According to different colored LP, these candidate regions may include LP regions; geometrical properties of LP are then used for classification. For the decomposing candidate regions, predetermined alphanumeric characters are used by position histogram to verify and detect vehicle LP regions. While conducting the experiment, vehicle images are taken under various conditions from traffic stations to evaluate the robustness, the flexibility and effectiveness.