License plate recognition based on mathematical morphology and template matching

License plate recognition systems usually include image preprocessing, license plate location and correction, character segmentation and character recognition. The system designed in this paper has studied and improved some of the above three steps. Firstly, the image preprocessing binarization of the license plate is carried out using the more adaptive otsu algorithm. Secondly, based on the mathematical morphology, the canny edge detection operator is used to realize the license plate location. The perspective transformation algorithm is used to correct the tilting problem of the license plate after positioning, and the vertical projection method is used to segment the corrected license plate image characters. Different from the traditional template matching algorithm to match pixels one by one, the Hausdorff distance algorithm is used to match the segmented license plate characters with the characters in the template library. Finally, the simulation results of matlab software show that the system can recognize the license plate characters of the existing license plate image.