Super Resolution to Identify License Plate Numbers in Low Resolution Videos

Abstract

Surveillance videos are useful as a source for extracting information in many areas such as crime investigations and monitoring offences in public roads. In many occasions, required information cannot be extracted from these videos due to the poor quality of the video or due to the large distance from camera to the interested object. A multiple image super resolution based technique to improve license plate regions in low quality videos has been proposed in this paper. Moreover, a two step image registration technique based on phase correlation method has been proposed to align multiple license plate image regions. The evaluation of the method was done based on an automatic number plate recognition system. The impact to the recognition rates of automatic number plate recognition system by using the proposed method is measured. Compressed, low quality, real traffic videos were used as the dataset. According to the results, a significant improvement of recognition rates achieved with the proposed method.