Evaluation of Super Resolution Methods to Increase Resolution of License Plate Images in License Plate Reader Systems

Reading license plate has applicable role in intelligent transportation system. In law enforcement systems and electronic toll collection systems, if reading license plate does not correctly done, then negative feedback from drivers can be provided. There is some reason for no detection of license plate like: low resolution license plate and small license plate because of distance. To overcome this drawback, the goal of this paper is to increase resolution of license plate image. Super resolution can be cheap software solution. In this paper based on image processing and intelligent training, novel methods for image super resolution are compared to select the best one for license plate image super resolution. Input is a low resolution image and output is higher resolution image correspond with low resolution image. These methods, with the help of many low resolution and high resolution images, enables the ability to increase license plate reader accuracy under low resolution conditions.