Design of License Plate Recognition System Based on Image Processing

In order to detect the character information of license plate in image quickly and conveniently, this paper designs a license plate recognition system based on image processing. The image processing operation in this system is based on the association function in OpenCV libraries. Firstly, the image is denoised by image preprocessing and morphological transformation, and then the license plate is located by contour search and filtering. Then, the extracted license plate continues to perform image processing to segment each character in the license plate. Finally, the convolutional neural network is used to train the model to recognize each character. The experimental results show that the system has good timeliness and recognition effect.