Low Visibility License Plate Area Detection Based on Dark Channel Prior Method and Top Hat Operation

People's travel and traffic safety are affected by low visibility. With the development of transportation in recent years, the detection of license plate area under low visibility has become one of the hot topics of research. This article studies the license plate detection under low visibility based on video images. The research content is divided into two parts: defogging and license plate inspection. First of all, the collected image is processed in grayscale, the related defogging coefficient is set, and the dark channel algorithm is used to remove the fog component in the image, which greatly improves the clarity of the image. Second, using the top hat calculation related theory, combined with the image edge extraction, opening and closing operation steps to detect the license plate area. Finally with the help of MATLAB software for test verification, the results show that the algorithm has a good detection effect.