A different approach for license plate recognition system

In this paper, we present a different approach for license plate recognition system which is a very popular subject in computer vision and pattern recognition fields. Generally, license plate recognition system is divided into three main parts: locating the license plate, character segmentation and character recognition. In locating the license plate phase which is the most important part of the license plate recognition system, we propose a new and original algorithm which uses the TR sign of the Turkish license plates after making horizontal scan for narrowing the search area. For segmenting the located license plate´s characters, we apply morphological operations to the binary image. At last, in the character recognition phase, we aim to minimize the error rate by separating the letter and number neural networks.