ADAPTIVE MULTI-NATIONAL LICENSE PLATE EXTRACTION
VLADIMIR SHAPIRO1, STEFAN BONCHEV2, VESELIN VELICHKOV2,
1 ORBOGRAPH LTD., P.O.BOX215, YAVNE 81102, ISRAEL,
2 INSTITUTE OF INFORMATION TECHNOLOGIES, 1113 SOFIA
Abstract: The paper represents the automatic plate localization component of a multi-national Car License Plate Recognition (CLPR) system. The approach includes stages of preprocessing, edge detection, filtering, detection of plate's position, slope evaluation, character segmentation and recognition. Single frame gray-level images are used as the only source of information. In the experiments Israeli and Bulgarian license plate images obtained at different daytime and weather conditions were used. The results have shown that the approach is robust to illumination, plate slope and scale and is insensitive to plate's country peculiarities.
Keywords: image processing, segmentation, license plate recognition.