Dual Thresholding Based License Plate Recognition

Applications that allow users to reserve parking spaces in advance, can incorporate license plate detection in a variety of ways. This may also be viewed as an effort to preserve order within the society. This feature enables the authorities (police) to track down the vehicle/person involved in the crime. This study suggests an algorithm for reading license plate text which includes steps such as RGB to gray scale conversion, thresholding of the pixel value, edge detection, contour detection, plate identification, text extraction. This paper suggests an algorithm which involves dual binarization methods: Otsu binarization method and binary thresholding thereby enhancing the accuracy of license plate detection.