License plate character segmentation and recognition method

Abstract

The invention discloses a license plate character segmentation and recognition method. The method comprises the following steps: S1, preprocessing a license plate image; S2, removing the spacer, removing the upper and lower frames, and positioning the coordinates of the right end of the second character; S3, segmenting the characters, and normalizing the characters; S4, dividing the acquired dataset into a training set and a test set; S5, constructing a convolutional neural network suitable for license plate character image recognition; S6, selecting training parameters, and training the setnetwork by using the training set; S7, testing the trained network by using the test set to obtain the accuracy of the license plate recognition network. According to the method, the problem of character segmentation failure caused by character breakage can be prevented, the overfitting problem caused by few training samples is solved, and the convergence rate and the generalization ability of themodel are further improved.

Description

technical field [0001] The invention relates to the technical field of license plate recognition, in particular to a method for character segmentation and recognition of a license plate. Background technique [0002] The license plate number is an important identification of the vehicle, so the license plate recognition technology is of great significance to the traffic control, and the license plate recognition system is an important link in the construction of an intelligent transportation system. [0003] Traditional character segmentation includes connected domain method and vertical projection method, which have a good effect on character segmentation, but when characters such as "chuan" and "zhe" are broken in the license plate, connected domain method and projection method will be wrong. The word "Chuan" is divided into three "1". [0004] The traditional template matching license plate recognition method relies on the matching degree between the template characters ...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to overcome the deficiencies of the prior art, to provide a method that can avoid the character segmentation failure problem caused by character breakage, solve the over-fitting problem caused by fewer training samples, and further improve the convergence speed and the generalization ability of the model license plate character segmentation and recognition method

Images