License Plate Location for Vehicles Passing through a Gate

This paper is dedicated to a license plate location method for moving vehicles passing through a gate in various weather conditions. The proposed license plate location algorithm mainly consists of edge detection, binarization, and license-plate judgment. Based on histogram analysis of a gradient image, the possible position of license plate is located from the image through the proposed effective edge detection. From those license-plate candidate regions, the correct position of the license plate is selected by the horizontal projection. In order to correctly extract the license plate in the sunny, cloudy and raining weathers, an adaptive binarization method is developed for substantially preserving edges of a license plate. Experimental results show that the proposed license plate location with low-complexity has good performance in locating the license plate of moving vehicles for various weathers. For vehicles passing through a gate at different speeds, the license plate location rates is above 90% on average for various weather conditions including the sunny situation, cloudy situation, and raining situation.