Real-time Automatic License Plate Recognition Using Color Features

Abstract

Various researchers presented various solutions for license plate detection but real-time performance is still a challenge in the field. In this paper, we propose a fast license plate detection method to work correctly in a real-time environment. In the first step, we locate or detect the license plate in the image sequences. We used color-based methods to detect the license plate. The method consists of computing image contours, later, we analyzed the contours to localize the license plate in the image sequences. After detecting the license plate, in the second step, we perform segmentation using a character recognition model. Finally, we propose the license plate format checking model to verify the detected license plate is the correct license plate. For the tools, we used OpenCV (open computer vision library) and tesseract for character recognition.