一种改进模板匹配的车牌字符识别方法

A license plate characters recognition method based on improved template matching

提出了一种改进模板匹配的车牌字符识别方法。详细介绍了车牌字符识别的过程,主要包括模板的建立,基于弹性网格的字符特征提取,模板匹配的步骤及改进方法。运用该方法不仅可减少计算量,而且提高了实时性。实验结果表明,提出的方法具有效率高、精确度好等特点,识别率可以达到90%以上。

This paper presents an improved template matching method of license plate characters recognition. License plate characters recognition process is introduced in detail, which includes the establishment of the template, the character feature extraction based on an elastic mesh, template matching steps and improved methods. By using this method, the computational amount is greatly reduced, while real-time is improved. The experimental result shows that the proposed method has high efficiency, accuracy and good characteristics, and the recognition rate can reach 90%.