面向车牌照识别数据的预处理方法研究

Research on Preprocessing Method for Vehicle License Plate Recognition Data

摘要: 为了实现对车牌照识别数据的有效利用，本文在分析车牌照识别数据特点和数据异常情况的基础上，结合数据挖掘技术中的数据预处理策略，设计了一种针对问题车牌照识别数据的预处理方法，该方法包括数据清理、数据集成、数据转换、数据归约4个过程。利用桂林市检测卡口的车牌照识别数据对该方法的有效性进行论证，结果表明该方法可以有效提升车牌照识别数据利用质量，降低数据异常对模型的影响。

Abstract: In order to distinguish the effective use of license plate recognition data, this paper analyzes the characteristics of vehicle license plate recognition data and data anomaly, combining data pre-processing strategies in data mining technology to design a vehicle license plate recognition data. This preprocessing method includes four processes of data cleaning, data integration, data con-version, and data reduction. The effectiveness of the method is demonstrated by using the license plate recognition data of the detection bayonet in Guilin. The results show that the method can improve the utilization quality of license plate identification data effectively and reduce the impact of data anomaly on the model.