Authors : Ali H.A. Aldabbagh, Laith A.H. Al-Shimaysawee and Hussein M.H. Al-Rikabi
Abstract: In this study, we present an algorithm to detect Iraqi car license plate and extract its number. Our method has two main stages. The first main stage is to detect the license plate which includes applying Sobel edge detector and morphological operators to the input image. After that the proposed license plates areas are compared to training images already recorded in the system using similarity index. The second main stage is the recognition process which is applied on the detected license plate; the number is extracted based on comparing each character with training numbers and letters already recorded in the system through image subtraction. The system performance is evaluated using significant number of images. The images were taken from different distances from to attain an accurate assessment. The proposed method showed superior performance at distances (1-3 m), however, as the distance increases the system accuracy decreases gradually.
Ali H.A. Aldabbagh, Laith A.H. Al-Shimaysawee and Hussein M.H. Al-Rikabi, 2019. Novel Algorithm for Iraqi Car License Plate Detection and Recognition. Journal of Engineering and Applied Sciences, 14: 205-210.