A Method of License Plate Location Based on Mathematical Morphology and Corner Detection

In order to reduce the effect of definition differences of the images monitored in all-weather conditions on speed and accuracy of the plate location, a fast license plate locating method is presented using texture and geometric features of license plate. Firstly, binary image is obtained by dynamic binarization. Then, the regions where gray values change frequently are marked initially. After a few mathematical morphology processing, plate region is located by corner detection and geometric features of license plate. Experimental results show that the method can locate the license plate of the image which monitored in all-weather conditions quickly and accurately, and accuracy rate can achieve 98.6% and the average locating time is 0.398s which prove the feasibility and validity of the method.