A Hybrid License Plate Extraction Method for Complex Scenes

This paper presents a hybrid method for extracting license plates from cluttered images. The proposed algorithm consists of three major components. First, a line detection method is proposed to detect straight lines in the edge map. Second, a weight assignment scheme is applied to obtain a weight based edge density map. Regions with the densest edges are selected as candidates. Third, all candidate plates will pass to a refining selection procedure. One candidate that best satisfy the color constraints becomes the output.