Locating car license plate under various illumination conditions using genetic algorithm

In this paper genetic algorithm is used for the license plate locating problem. To overcome the degradation of license plates in vehicle images under various illumination conditions, a windowed local area is adaptively binarized and the feature vector is obtained from a bank of filter. After that, the real integer coded genetic algorithm is proposed to search the whole image for the most promising area. Experimental results show that the local area processing combined with GA is quite robust with images of various illumination conditions compared with other techniques.