A Character Recognition Method for License Plate Based on Combination Feature and Partial Character Input

摘要

In order to improve the recognition rate of license plate characters, a new method based on combination feature and partial character input is proposed. First, according to the texture characteristics of the license plate character, the basic LBP operator is improved, and then the characteristics of the license plate partial character image are extracted using the improved local binary mode and horizontal vertical projection, and the classifier is trained with the extracted features. Then use this classifier to classify and identify license plate characters. The experimental results show that the overall recognition rate of license plate reaches 94.56 %, which proves the effectiveness and robustness of this method.