License plate character segmentation based on horizontal projection and connected component analysis

In this paper, license plate segmentation system is proposed for Algerian vehicles. The proposed system is divided into two main parts: identification of the license plate from the input image and then segmented the characters from license plate. After pre-processing the input image, first, a new method is used to locate the license plate at the right position within the acquired image by using horizontal projection. Then, the license plate image is segmented using connected component labeling algorithm. The results showed that the proposed method is robust to the different lighting condition, image deterioration, background interference and so on, with 96.3% of accuracy rate.